Clinical Practice Guideline for the Treatment of Intrinsic Circadian

Home , Chronotype, Circadian rhythm, Dark therapy, Light therapy, Phase response curve, Sleep

SPECIAL ARTICLES

pii: jc-00358-15 http://dx.doi.org/10.5664/jcsm.5100 Clinical Practice Guideline for the Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders: Advanced Sleep- Wake Phase Disorder (ASWPD), Delayed Sleep-Wake Phase Disorder (DSWPD), Non-24-Hour Sleep-Wake Rhythm Disorder (N24SWD), and Irregular Sleep-Wake Rhythm Disorder (ISWRD). An Update for 2015 An American Academy of Sleep Medicine Clinical Practice Guideline R. Robert Auger, MD1; Helen J. Burgess, PhD2; Jonathan S. Emens, MD3; Ludmila V. Deriy, PhD4; Sherene M. Thomas, PhD4; Katherine M. Sharkey, MD, PhD5 1Mayo Center for Sleep Medicine, Rochester, MN; 2Rush University Medical Center, Chicago, IL; 3Portland VA Medical Center, Portland, OR; 4American Academy of Sleep Medicine, Darien, IL; 5Brown University, Providence, RI

A systematic literature review and meta-analyses (where second- and fi rst-tier degree of confi dence, respectively. No appropriate) were performed and the GRADE approach was recommendations were provided for remaining treatments/ used to update the previous American Academy of Sleep populations, due to either insuffi cient or absent data. Areas Medicine Practice Parameters on the treatment of intrinsic where further research is needed are discussed. circadian rhythm sleep-wake disorders. Available data Keywords: circadian rhythms, DSWPD, ASWPD, N24SWD, allowed for positive endorsement (at a second-tier degree of ISWRD confi dence) of strategically timed melatonin (for the treatment Citation: Auger RR, Burgess HJ, Emens JS, Deriy LV, of DSWPD, blind adults with N24SWD, and children/ Thomas SM, Sharkey KM. Clinical practice guideline for the adolescents with ISWRD and comorbid neurological disorders), treatment of intrinsic circadian rhythm sleep-wake disorders: and light therapy with or without accompanying behavioral advanced sleep-wake phase disorder (ASWPD), delayed interventions (adults with ASWPD, children/adolescents with sleep-wake phase disorder (DSWPD), non-24-hour sleep- DSWPD, and elderly with dementia). Recommendations wake rhythm disorder (N24SWD), and irregular sleep-wake against the use of melatonin and discrete sleep-promoting rhythm disorder (ISWRD). An update for 2015. J Clin Sleep medications are provided for demented elderly patients, at a Med 2015;11(10):1199– 1236.

SUMMARY the TF. The TF developed consensus-based relevant outcomes, rated their relative importance, and determined clinical signifi - cance thresholds. Extracted data were pooled across studies Purpose for each outcome measure in accordance with PICO questions, The present document replaces/updates the previous Ameri- and based upon CRSWD diagnosis, study design, patient pop- can Academy of Sleep Medicine (AASM) Practice Parameters ulation, outcome of interest, and method of derivation. Statis- pertaining to the intrinsic CRSWDs (i.e., ASWPD, DSWPD, tical analyses were performed using dedicated software, and N24SWD, and ISWRD). The treatment of remaining CRSWDs meta-analyses were completed when applicable. The GRADE is not addressed. (Grading of Recommendations Assessment, Development, and Evaluation) approach was used to develop recommenda- Methodology tion statements and to determine the direction and strengths of The AASM commissioned a Task Force (TF) of 4 members these recommendations based upon a composite assessment of with expertise in the fi eld of CRSWDs, appointed a Board of evidence quality, benefi ts versus harms analyses, and patient Directors (BOD) liaison, and assigned a Science and Research values and preferences. Department staff member to manage the project. PICO (Pa- tient, Population or Problem, Intervention, Comparison, and Findings Outcomes) questions were developed by the TF and approved Available data allowed for positive endorsement (at a sec- by the BOD. Extensive literature searches were performed to ond-tier degree of confi dence) of strategically timed melatonin identify articles of interest, and relevant data were extracted by (for the treatment of DSWPD, blind adults with N24SWD, and

1199 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al. children/adolescents with ISWRD and comorbid neurologi- 5.4.6.2a The TF suggests that clinicians use strategically cal disorders), and light therapy with or without accompany- timed melatonin as a treatment for ISWRD in children/ ing behavioral interventions (adults with ASWPD, children/ adolescents with neurologic disorders (versus no treatment). adolescents with DSWPD, and elderly with dementia and [WEAK FOR] ISWRD). Recommendations against the use of melatonin and discrete sleep-promoting medications are provided for de- 5.4.9.1a The TF suggests that clinicians avoid the use mented elderly patients, at a second- and first-tier degree of of combined treatments consisting of light therapy in confidence, respectively. No recommendations were provided combination with melatonin in demented, elderly patients for remaining treatments/populations, due to either insufficient with ISWRD (versus no treatment). [WEAK AGAINST] or absent data. Conclusion Recommendations are as Follows Use of the GRADE system for this updated Clinical Prac- tice Guideline represents a major change. This update should ASWPD provide clinicians with heightened confidence with respect to 5.1.4a The TF suggests that clinicians treat adult ASWPD prescribing select treatments and, equally importantly, should patients with evening light therapy (versus no treatment). serve as a roadmap for future studies that will propel higher [WEAK FOR] quality, more sophisticated therapies for the intrinsic CRSWDs.

DSWPD 5.2.6.1a The TF suggests that clinicians treat DSWPD in adults with and without depression with strategically timed melatonin (versus no treatment). [WEAK FOR]

5.2.6.2.1a The TF suggests that clinicians treat children and adolescents with DSWPD (and no comorbidities) with strategically timed melatonin (versus no treatment). [WEAK FOR]

5.2.6.2.2a The TF suggests that clinicians treat children and adolescents with DSWPD comorbid with psychiatric conditions with strategically timed melatonin (versus no treatment). [WEAK FOR]

5.2.9.2a The TF suggests that clinicians treat children and adolescents with DSWPD with post-awakening light therapy in conjunction with behavioral treatments (versus no treatment). [WEAK FOR]

N24SWD 5.3.6.1a The TF suggests that clinicians use strategically timed melatonin for the treatment of N24SWD in blind adults (versus no treatment). [WEAK FOR]

ISWRD 5.4.4a The TF suggests that clinicians treat ISWRD in elderly patients with dementia with light therapy (versus no treatment). [WEAK FOR]

5.4.5a The TF recommends that clinicians avoid the use of sleep-promoting medications to treat demented elderly patients with ISWRD (versus no treatment). [STRONG AGAINST]

5.4.6.1a The TF suggests that clinicians avoid the use of melatonin as a treatment for ISWRD in older people with dementia (versus no treatment). [WEAK AGAINST]

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1200 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

1.0 INTRODUCTION for the governance of sleep and wakefulness: “Process S” and “Process C.”6 The homeostatic drive to sleep (Process S) is pro- The American Academy of Sleep Medicine (AASM) pro- portional to the duration of wakefulness. In contrast, Process C duced the first Practice Parameters (and associated reviews) for creates a drive for wakefulness that variably opposes Process S the evaluation and treatment of circadian rhythm sleep-wake and is dependent upon circadian (“approximately daily”) rhythms disorders (CRSWDs) in 2007.1–3 The purpose of the present intrinsic to the individual. Master coordination of this sleep/wake publication is to provide an evidence-based update of existing rhythm is provided by the neurons of the suprachiasmatic nuclei recommendations for the treatment of the intrinsic CRSWDs: located within the hypothalamus.7–10 As this intrinsic period is advanced sleep-wake phase disorder (ASWPD), delayed sleep- typically slightly longer than 24 hours in humans11,12 synchroni- wake phase disorder (DSWPD), non-24-hour sleep-wake zation to the 24-hour day (entrainment) is accomplished by vari- rhythm disorder (N24SWD), and irregular sleep-wake rhythm ous environmental inputs, the most important of which is light disorder (ISWRD). The extrinsic or predominantly environ- and dark exposure.13 Failure to synchronize can alter the phase mentally influenced CRSWDs, namely shift work and jet lag relationships between internal rhythms and the light/dark cycle, disorder, are not addressed in this paper. which may manifest in the form of circadian rhythm sleep-wake disorders (CRSWDs). The intrinsic CRSWDs refer to those con- 2.0 BACKGROUND ditions that are thought to exist predominantly due to innate phe- nomena, although exogenous components contribute to varying Reviewed studies that included patients with an explicitly degrees in all of these disorders. stated CRSWD predominantly utilized the International Classi- The intrinsic CRSWDs are briefly characterized as follows. fication of Sleep Disorders, Second Edition (ICSD-2)4 diagnostic DSWPD manifests as a delay of the major sleep episode with criteria, despite the fact that International Classification of Sleep respect to the patient’s desired timing or the timing required to Disorders, Third Edition (ICSD-3)5 nomenclature is referenced attend to social, educational, and/or occupational demands. Pa- throughout the manuscript. Important modifications to the In- tients report extreme difficulty both with falling asleep at bed- ternational Classification of Sleep Disorders include incorpora- times considered typical among their peers, and with waking at tion of the word “wake” (the ICSD-2 referred solely to circadian the required or desired times, but sleep quality is typically re- rhythm sleep disorders), which highlights the significant impair- ported as normal when the individual sleeps at the delayed times. ments these conditions exert on daytime functioning. Caregiver In contrast, an advance of the major sleep episode with respect input is also emphasized in the ICSD-3, particularly with respect to the patient’s desired or required sleep-wake times character- to diagnostic assessments among cognitively impaired and pedi- izes ASWPD. ASWPD patients report extreme difficulty staying atric patients. Other major changes include the recommendation awake during evening hours and frequently note falling asleep that CRSWD diagnoses are ascertained via actigraphy derived before completion of pertinent work, social, or family obligations. data when possible (with inclusion of both work/school and In addition, wake time is undesirably early, and considered atypi- free days), to provide objective longitudinal documentation of cal in comparison to peers. For both conditions, symptoms must sleep-wake patterns. Consistent with this emphasis on objective be present for at least 3 months and schedules need to be docu- measures, circadian phase assessments (e.g., dim light melato- mented with sleep diaries and/or wrist actigraphy for a period of nin onset, or DLMO) are also recommended, if feasible. Other at least 7 days. changes include a de-emphasis on “conventional” and “socially N24SWD is diagnosed when patients fail to entrain to the acceptable” clock times (recognizing the relative nature of these 24-hour light-dark cycle and clock times. Thus, patients exhibit terms, and instead highlighting patients’ subjective concerns), sleep-wake patterns that show a progressive delay (usually) or extensive additions to the “Pathology and Pathophysiology” and advance, depending upon the period length (tau) of their en- “Polysomnographic and Other Objective Findings” sections, and dogenous circadian rhythms. During a symptomatic period, the new descriptions of “Developmental Issues” and “Clinical and time of high sleep propensity gradually shifts, such that patients Pathophysiologic Subtypes.”5 experience daytime hypersomnolence and nighttime insomnia. In many instances, this review incorporated trials with par- Most patients with N24SWD are totally blind, but this disorder ticipants who were not recruited in strict accordance with In- also occurs among sighted individuals. In contrast to the other ternational Classification of Sleep Disorders criteria, but who CRSWDs, a N24SWD diagnosis requires at least 14 days of doc- nonetheless described symptoms consistent with a CRSWD umentation of progressively shifting sleep-wake times with sleep (based upon Task Force consensus). Examples include pediat- diaries and/or actigraphy. ric/adolescent patients with “idiopathic sleep-onset insomnia,” Patients with ISWRD lack a clear circadian pattern of sleep- whose symptoms were consistent with DSWPD, as well as wake behavior. Thus, afflicted individuals experience pro- select populations of institutionalized elderly patients, among longed periods of wakefulness during the typical nocturnal whom varied descriptions of insomnia, nighttime wakefulness, sleep episode in addition to excessive sleepiness and prolonged and daytime napping selectively appeared to be representative sleep bouts during daytime hours. Sleep is fragmented and of ISWRD, despite the fact that this condition was not named frequently insufficient. ISWRD is observed more commonly explicitly. A similar approach was taken for this latter group of among patients with neurodevelopmental or neurodegenera- patients during literature review and development of the previ- tive disorders, and can pose particular challenges for caregiv- ous Practice Parameters.1,3 ers. Documentation (sleep diaries and/or actigraphy) of multiple A brief scientific background is required. The two-process non-circadian sleep-wake bouts for a period of at least 7 days is model for sleep regulation delineates two principle mechanisms required for diagnosis.

1201 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

Table 1—PICO question parameters. Population Intervention Comparison Outcomes Patients diagnosed 1. Prescribed sleep-wake scheduling Control group, those Physiologic circadian phase with intrinsic 2. Timed physical activity/exercise treated with placebo markers (DLMO [saliva/plasma], CRSWDs 3. Strategic avoidance of light (e.g., with the use of eyewear) or, when a comparison urinary melatonin metabolite,

(ASWPD, DSWPD, 4. Light therapy group was not available, constant routine CBTMin) N24SWD, ISWRD) 5. Sleep-promoting medications (hypnotics/sedatives/ measurements performed Total sleep time (TST) neuroleptics/other novel agents) “before” (baseline) and 6. Timed oral administration of melatonin or agonists “after” treatment Initial sleep latency (ISL) 7. Wakefulness-promoting medications (e.g., modafinil, Sleep onset time (SOT) traditional stimulants) 8. Other somatic interventions Sleep offset time (SOffT) 9. Combination treatments

Interventions for CRSWDs can be broadly categorized as fol- a Task Force (TF) of four members with expertise in the field of lows: (1) prescribed timing of sleep-wake and/or physical activity/ CRSWDs, appointed an AASM Board of Directors (BOD) liai- exercise, (2) strategic receipt and/or avoidance of light, (3) use of son, and assigned an AASM Science and Research Department medications and/or supplements to phase shift and/or to promote staff member to manage the project. Prior to appointment, the sleep or wakefulness, and (4) alternate interventions that exert ef- content experts were required to disclose all potential conflicts fects by altering bodily functions to impact sleep/wake behaviors of interest according to AASM policy. None were declared. The (i.e., somatic interventions). TF performed an extensive review of the scientific literature Light is strategically timed according to phase response and assessed the available evidence employing the methodol- curves (PRCs).2 In brief, light can suppress melatonin secretion14 ogy of evidence-based medicine (specifically, meta-analysis and phase shift circadian timing in humans,15 leading to the use and the Grading of Recommendations Assessment, Develop- of timed light exposure as a treatment for CRSWDs. Light timed ment and Evaluation system, or GRADE) to draft recommenda- in the evening and before the core body temperature minimum tions. The present paper was approved by the AASM BOD and 1 (CBTmin) leads to phase delays, and light timed after the CBT- replaces the previous Practice Parameters. The AASM expects 15 min in the morning leads to phase advances. Larger effects are these guidelines to have a positive impact on clinical decision- observed with greater intensities of light and longer durations making and patient outcomes. These recommendations reflect of light, but the increases are nonlinear.16,17 Additionally, the re- the state of knowledge at the time of publication and will be sponse to light is modified by prior exposure to light or “light revised when the availability of new information necessitates. history,”18,19 such that a history of less light exposure leads to a greater response to light. Just as light exposure can shift circa- 3.2 PICO Questions dian timing, so too can the strategic avoidance or reduction of Eight PICO (Patient, Population or Problem, Intervention, light.20,21 Finally, the human circadian system is most sensitive Comparison, and Outcomes) questions were developed, based to short wavelength blue light (~480 nm),22,23 although at bright on both the inquiries raised in the previous AASM publica- intensities phase shifts to white broad spectrum light and blue tions1,3 and an investigation of systematic reviews, meta-anal- enriched light are similar, presumably due to a saturation of yses, and guidelines published subsequently (Table 1). The photoreceptors.24,25 AASM BOD ultimately approved these questions. In addition, Less is known about the variables contributing to melatonin combination treatments were also reviewed for the four intrin- response.2 The melatonin PRC is approximately 180 degrees out sic CRSWDs included in this guideline. of phase with the light PRC, such that dosing in the afternoon/ evening shifts rhythms earlier and dosing in the morning shifts 3.3 Literature Searches rhythms later. As the CBTmin serves as the “inflection point” be- Literature search #1 was performed in PubMed using broad tween delaying and advancing effects for light, the DLMO serves terms (see Appendix), in order to identify systematic reviews, as the approximate inflection point for advancing and delaying meta-analyses or relevant practice guidelines published sub- effects of melatonin. Optimal dosing of melatonin for circadian sequent to availability of the previous AASM Practice Pa- effects remains unclear, and studies suggest that timing is more rameters. Examination of discovered papers (n = 93) enabled important than dose (PRCs for doses above 5 mg have not been elucidation of Practice Parameter recommendations requir- published). In addition to phase shifting effects, melatonin may ing revisions, and also assisted with further refinement of the also have direct soporific effects, particularly at higher doses. PICO questions. The next literature search (#2) targeted treatment trials involving intrinsic CRSWDs that addressed at least one PICO question. This search utilized PubMed, Embase, and 3.0 METHODS PsycInfo databases. At least two TF members carefully assessed the abstract of 3.1 Expert Task Force each retrieved article (n = 2,063), to determine whether the In order to develop these Clinical Practice Guidelines, the publication should be included for further consideration. The American Academy of Sleep Medicine (AASM) commissioned following list of general exclusion criteria was used:

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1202 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

Table 2—CRITICAL outcomes and their TF-defined clinical significance thresholds. Clinical Significance Thresholds Circadian Phase TST (change in ISL (change in SOT (change in SOffT (change in Entrainment Diagnosis (change in minutes) minutes) minutes) minutes) minutes) Status ASWPD 30 30 15 15 15 N/A DSWPD 30 30 15 15 15 N/A ISWRD 30 30 15 15 15 N/A N24SWD N/A N/A N/A N/A N/A Yes/No

1. Diagnosis or not treatment rated their relative importance. Physiologic circadian phase 2. Not CRSWD markers, total sleep time (TST), initial sleep latency (ISL), 3. Not intrinsic CRSWD (shift work or jet-lag disorder) sleep onset time (SOT), and sleep offset time (SOffT) were 4. Wrong publication type (review, editorial, etc.) deemed CRITICAL for making recommendations, and a sig- 5. Not human subjects nificance threshold was defined for each outcome based upon 6. Mechanistic or methodological study consensus (Table 2), as no published standards exist. Sleep 7. Study was published before October 2006 parameters were alternately evaluated with polysomnogra- When there were questions or disagreements, the full text phy (PSG), wrist actigraphy, or sleep diaries. Although both of the article was reviewed in detail until consensus was wakefulness after sleep onset and sleep efficiency were also reached. The same search terms, databases and inclusion/ex- commonly reported, the two variables were rated as IMPOR- clusion criteria were used for literature search #3, although TANT (but not CRITICAL) by the TF. As such, related data new date limitations were applied (June, 2012–March, 2014), did not factor into clinical recommendations. A unique sce- with the intention to capture articles published after comple- nario arose for N24SWD, for which entrainment status (i.e., tion of search #2. Four hundred fifty-three additional publica- whether the biological clock is synchronized to the 24-hour tions were retrieved, and TF assessments occurred in the same day) was solely utilized as a CRITICAL outcome measure, manner described above. Finally, TF members selected several as it defines this CRSWD physiologically (see section 5.3 in literature reviews (by consensus), and screened reference lists this paper). to identify other articles of potential interest. This “pearling” process served as a “spot control” for the previous searches, 3.5 Extraction of Evidence and ensured that important articles were not missed. All dupli- Quantitative data pertaining to the outcomes of interest as cate references were eliminated. well as information necessary for systematic evaluation and Since new inclusion/exclusion criteria were used in this grading of the evidence were extracted from accepted articles project, investigations cited in the previous Practice Param- using a dedicated spreadsheet. Articles determined to lack eters1 were not necessarily incorporated into the current analy- quantitative data or with data presented in a format incompat- sis. Studies that did not meet inclusion criteria were selectively ible with desired statistical analyses were rejected at this stage, used for discussion purposes, but were neither included in the but used selectively for further discussion. In instances where GRADE reports nor used as a basis for recommendations. The desired data were available but not presented in the desired for- TF made a particular effort to discuss those studies (contain- mat, the authors were contacted, and raw data were acquired if ing either patients or healthy subjects) that might spur and/or possible. Data were pooled across the studies for each outcome improve future clinical research for the reviewed CRSWDs. measure in accordance with PICO questions and based on A final PubMed search was conducted to identify harms diagnosis, study design, patient population, clinical outcome or adverse effects attributed to the relevant interventions: of interest, and method of derivation (e.g., PSG derived data light therapy (PICO 4), hypnotics (PICO 5), and melatonin were analyzed separately from data derived from actigraphy (PICO 6) (see Appendix). Limitations were imposed to select or sleep diaries). for English-language “meta-analyses” and “systematic reviews” pertaining to human subjects. The titles and abstracts 3.6 Statistical Analyses of articles produced by these searches were reviewed for rel- Meta-analyses were completed (in the few instances pos- evance, and pertinent publications were examined. Other cited sible) using the random effects model. All computations were articles from the “Harms and Adverse Effects” section were performed using the Review Manager software,26 and included culled from prior searches (but deemed ineligible for quantita- calculations of the mean difference (MD) ± standard deviation tive analysis) or were provided via TF members’ preemptive (SD) for CRITICAL outcomes. Values analyzed in this manner awareness and consensus regarding relevancy. Adverse effects are displayed to the hundredths place. Age demographics and of combined treatments were addressed based on the singular information regarding melatonin doses are presented in the components of combinations. format provided by the associated study (mean ± SD if available) but, in an effort to maintain consistency, are displayed 3.4 Treatment Efficacy Outcomes only to the tenths place in instances where the authors pro- During the process of data extraction, the TF developed vided values with numerical place values of lower hierarchy. a list of patient-oriented clinically relevant outcomes and The results of meta-analyses are depicted in figures within the

1203 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

Figure 1—Guide for interpretation of clinical significance of the results. A B C ne ne ne i i i l l l t t t c c c e e e f f f f f f e e e o o o N N N

Clinical significance Clinical significance Clinical significance threshold defined by the TF threshold defined by the TF threshold defined by the TF

Examples of (A) clinically significant improvement;(B) “serious” imprecision, grade one level down; (C) “very serious” imprecision, grade two levels down. text, in association with a “forest plot.” Summary of findings both sides of the no effect line, the evidence is graded two lev- tables for all investigations are presented in the Appendix. els down for “very serious imprecision” (Figure 1C). Since When studies contained placebo/control groups, the evalu- the Review Manager software does not operate with clinical ation of the effect of treatment was performed by comparison significance thresholds, these dotted lines are not depicted in of averaged posttreatment and averaged post-placebo/control the figures associated with the actual data. The interpretation group values, regardless of the authors’ approaches. In studies of clinical significance from results of individual studies was with crossover or “before-after” designs where there was no accomplished in the same manner, but forest plots were not placebo/control group, posttreatment values were compared to created. baseline values. Our use of this methodology occasionally produced results that differed from those reported in the original 3.8 Quality of Evidence publications (e.g.27–29). The GRADE approach was used for the assessment of quality of evidence30–37 (also see: http://www.gradeworkinggroup. 3.7 Interpretation of Clinical Significance of Results org/publications/JCE_series.htm). In contrast to other meth- Interpretation of clinical significance was ascertained via ods, an estimate of effect is generated for critical outcomes comparisons with predefined thresholds (see Table 2 and across studies, as opposed to an evaluation of individual stud- Figure 1). For meta-analyses, the pooled MD (black diamond) ies. Multiple aspects of quality of evidence are assessed, with on “forest plots” depicts the average response or magnitude of downgrading of evidence as applicable (see Table 3). effect across all studies, the width of the diamond represents GRADE assigns high quality to evidence from randomized the associated 95% confidence interval (CI), and the “no ef- controlled trials while evidence from observational studies is fect” line represents nil benefit from the intervention. The dot- considered low quality. However, high quality evidence can ted lines on the left and right sides (equidistant from the “no be graded down, and low quality evidence can be graded up, effect” line) represent clinical decision thresholds defined by based upon the factors described below (see Table 3). The the TF (Figures 1A, 1B, and 1C). The right side represents an analysis of risk of bias includes review of the presence/ab- increase in the outcome measure, while the left represents a sence of blinding, allocation concealment, loss to follow up, decrease. An increase in some outcome measures, such as TST, or selective outcome reporting. Indirectness occurs when represents improvement. If the black diamond of TST data lies the question being addressed is different than the available beyond the clinical significance threshold on the right side, the evidence in terms of population, intervention, comparator, or result of a treatment is interpreted as a clinically significant outcome. There is inconsistency when there is unexplained improvement (Figure 1A). If, however, the diamond lies to the heterogeneity of the results. Imprecision is described in sec- left of the “clinical significance” line, the decrease is regarded tion 3.7 in this paper. as a clinically significant undesired outcome, and the treatment In GRADE, there are 4 specific categories for assessing the is deemed contraindicated. When improvement is signified by quality of a body of evidence. a decrease in the outcome measure (e.g., ISL), the interpreta- High: corresponds to a high level of certainty that the true tion is reversed. effect lies close to that of the estimate of the effect. When the confidence interval crosses the clinical signifi- Moderate: corresponds to a moderate level of certainty in cance threshold on one side, the evidence is graded one level the effect estimate; the true effect is likely to be close to down (Figure 1B) for “serious imprecision.” When the con- the estimate of the effect, but there is a possibility that it fidence interval crosses the clinical significance threshold on is substantially different.

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1204 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

Table 3—Summary of GRADE approach to rating quality of evidence (from Guyatt et al.30). Initial Quality of a Study Design Body of Evidence Downgrade if Upgrade if Quality of a Body of Evidence Randomized trials High → Risk of bias Large effect High (four plus: ) −1 Serious +1 Large −2 Very serious +2 Very large ⊕⊕⊕⊕ Inconsistency Dose response −1 Serious +1 Evidence of a Moderate (three plus: ) −2 Very serious gradient ⊕⊕⊕⊝ Indirectness All plausible residual Observational studies Low → −1 Serious confounding Low (two plus: ) −2 Very serious +1 Would reduce a demonstrated ⊕⊕⊝⊝ Imprecision effect −1 Serious +1 Would suggest −2 Very serious Very Low (one plus: ) a spurious Publication bias effect if no ⊕⊝⊝⊝ −1 Serious effect was −2 Very serious observed

Low: corresponds to a low level of certainty in the effect including discussion in the referenced papers about estimate; the true effect may be substantially different tolerability, compliance, and patients’ experiences with from the estimate of the effect. the treatments in question, the TF judged whether: Very low: corresponds to very little certainty in the effect a. The vast majority of well-informed patients (> 90%) estimate; the true effect is likely to be substantially would most likely use this patient-care strategy, different from the estimate of effect. compared to alternative patient-care strategies or no The body of evidence for each outcome was assessed and treatment graded, taking into account quality considerations based on b. The majority of well-informed patients would most the quantitative analysis and other major factors described likely use this patient-care strategy, compared to above. CRITICAL outcome results are presented as summary alternative patient-care strategies or no treatment of findings tables organized by PICO question and patient pop- c. The majority of well-informed patients would most ulation (see Appendix, Tables S1–S12). likely NOT use this patient-care strategy, compared A cumulative quality grade for a particular PICO question to alternative patient-care strategies or no treatment and patient population is predicated upon the lowest level of d. The vast majority of patients (> 90%) would most evidence assigned to one of the CRITICAL outcomes. Thus, if likely NOT use this patient-care strategy, compared a recommendation was based upon two outcomes, one of mod- to alternative patient-care strategies or no treatment erate and one of low quality, the overall grade would be low. Taking these variables into consideration, each recommendation statement was given a “strength value” of Strong For, 3.9 Strength of Recommendations Weak For, Weak Against or Strong Against (see Table 4). As The TF developed recommendation statements and deter- an example, a body of evidence could be rated VERY LOW, mined the strengths of these recommendations based on the while another could be rated MODERATE, and yet both could balance of the following major factors: generate a WEAK FOR recommendation, based upon the 1. Level of evidence—based on an assessment of the above mentioned factors. quality of evidence using GRADE criteria (see Table 3), There were multiple cases when the TF chose to make “No the TF categorized the evidence as: Recommendation,” which reflects either a complete lack of a. High available evidence (no studies were published) or situations b. Moderate when evidence was available but either did not meet review c. Low inclusion criteria or was considered insufficient to support a d. Very Low recommendation (see Appendix, Tables S5 and S6). It needs 2. Benefits vs. Harms—based upon CRITICAL outcomes to be emphasized that “No Recommendation” does not mean and analysis of any harms/side effects, the TF assessed that treatments should not be tried but that, in the absence of whether: sufficient and conclusive evidence, clinicians should use their a. Benefits outweighed harms best judgment to decide in each particular case whether or not b. Benefits equaled harms a treatment should be used. At the step of review of the ex- c. Harms outweighed benefits tracted evidence, the TF made a decision to exclude studies d. The balance between benefits and harms was unclear with fewer than 10 subjects if the study constituted a single 3. Patient values and preferences—based on the clinical source of evidence, as it was felt that affiliated data were insuf- expertise of the TF and relevant published data, ficient to support a recommendation.

1205 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

Table 4—Definitions of AASM strengths of recommendations. AASM Strength of Recommendation Example Characteristics Guiding Recommendation • There is a high degree of clinical certainty in the net benefits of this patient-care strategy. • The vast majority of well-informed patients would most likely choose this patient-care strategy, compared to STRONG alternative patient-care strategies or no treatment. FOR • There is a lower degree of clinical certainty in the balance between benefits vs. harms (e.g., net benefits) of this patient-care strategy. WEAK • The majority of well-informed patients would most likely choose this patient-care strategy, compared to alternative patient-care strategies or no treatment. • There is a lower degree of clinical certainty in the balance between benefits vs. harms (e.g., net harms) of this patient-care strategy. WEAK • The majority of well-informed patients would most likely not choose this patient-care strategy, compared to alternative patient-care strategies or no treatment. AGAINST • There is a high degree of clinical certainty in the net harms of this patient-care strategy. • The vast majority of well-informed patients would most likely not choose this patient-care strategy, STRONG compared to alternative patient-care strategies or no treatment.

associated with a lack of reported serious adverse effects.44–48 A 4.0 HARMS AND ADVERSE EFFECTS review by the National Academy of Sciences stated that short- term use of ≤ 10 mg/daily (higher than typical chronobiotic 4.1 Light Therapy doses) appears to be safe in healthy adults but recommended No studies/reviews were identified that specifically ad- caution in children/adolescents and women of reproductive dressed potential harms among patients with CRSWDs. In age (see further below). Adverse effects such as headaches, their Cochrane Systematic Review for the treatment of non- somnolence, hypotension, hypertension, gastrointestinal up- seasonal depression, Tuunainen and colleagues38 found that set, and exacerbation of alopecia areata have been reported hypomania was the sole side effect that was more common at higher melatonin doses in healthy adults, and the same among patients receiving light therapy versus controls (Rela- effects have been reported at lower doses among those with tive Risk 4.91 [CI 1.66–4.46]). Nevertheless, light treatment relevant preexisting conditions. Melatonin has also been asso- has been safely used for the treatment of bipolar depression, ciated with an increase in depressive symptoms,49 and caution with careful monitoring.39 is advised when prescribing to patients taking warfarin and to Other commonly described side effects include eyestrain, patients with epilepsy, as a result of various case reports sub- nausea, and agitation, albeit with predominant spontaneous mitted to the World Health Organization.45 A recent publica- remission. Treatment-emergent headaches also commonly re- tion described impairment in glucose tolerance among healthy mit,40 but light therapy can induce migraines in approximately women50 subsequent to acute melatonin administration. one-third of those susceptible.41 Studies that address long-term effects are scarce, as are stud- Finally, although commercially available products do not ies that specifically involve pediatric/adolescent populations. emit ultraviolet light, patients with eye disease and/or those A randomized, placebo-controlled trial that investigated the using photosensitizing medications should only use light toxicology of a 28-day treatment with 10 mg melatonin (solely therapy with periodic ophthalmological and/or dermatologi- comprised of healthy male adult participants) revealed no cal monitoring of the underlying condition.40,42,43 Reassuringly, group differences with respect to adverse effects on polysom- one study reported no changes in extensive ophthalmologic ex- nographically recorded sleep, subjective sleepiness, numerous aminations among seasonal affective disorder patients without clinical laboratory examinations, or other subjectively recorded preexisting conditions after up to 6 years of daily use in the fall events.51 Similarly, in a meta-analysis that reviewed controlled and winter months.42 trials with melatonin (n = 10 studies, over 200 subjects) used for ≤ 3 months, there were few reports of adverse events.48 4.2 Melatonin A long-term follow-up study of pediatric patients with Melatonin is considered a dietary supplement, and is there- DSWPD + attention deficit hyperactivity disorder (ADHD) fore not subject to the scrutiny afforded to United States Food who utilized melatonin doses up to 10 mg (mean follow-up time and Drug Administration (FDA)-approved medications. Con- of approximately 4 years) detected no serious adverse events cerns have been raised about the purity of available prepara- as a result of serial interviews with the children’s parents, and tions, as well as the reliability of stated doses. Formulations 65% of participants continued to use the medication daily.52 A that are United States Pharmacopeial Convention Verified can follow-up open-label prospective study of subjects with neu- be considered most reliable in this regard. rodevelopmental disabilities comorbid with DSWPD who re- Few identified papers addressed risks or side effects specifi- ceived controlled-release melatonin (max dosage 15 mg) up to cally among patients with CRSWDs. In general, melatonin is 3.8 years similarly described no adverse events.53,54 Patients and

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1206 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders caregivers are nevertheless frequently wary to use this supple- and offset typically occur two or more hours prior to required ment, due to concerns related to potential adverse effects on or desired sleep times. Affected individuals complain of early growth hormone regulation (10 mg dose),55 and on reproductive morning or maintenance insomnia and excessive evening function/development (3 mg dose).56 Possibly relevant to the sleepiness. When allowed to maintain an advanced schedule, latter concern, Tanner stages were assessed serially in a ques- sleep quality and quantity are improved.5 tionnaire-based study involving children/adolescents (mean duration ~3 years), in an effort to compare pubertal development 5.1.1 Prescribed Sleep-Wake Scheduling for Patients with among those using melatonin (mean dose ~3 mg) during pre- ASWPD puberty to non-melatonin users in the general Dutch population Sleep-wake scheduling has only been described in one case (controls).57 No significant group differences were detected. report, but favorably affected the predefined CRITICAL outcomes (SOT and SOffT, both subjective) in ASWPD.67 No new 4.3 Hypnotics studies were identified. General adversities attributed to sedative-hypnotics (though In the report of a 62-year-old male, sleep times were ad- not specifically among patients with CRSWDs) include in- vanced 3 hours every 2 days for 2 weeks and then stabilized at creased risks for falls, headaches, nausea, medication-med- the desired times, which were maintained at 5 months follow- ication interactions, and drug dependence,58,59 with elderly up. This treatment was designated as an OPTION in the 2007 patients at specific high risk.60–62 Data regarding the use of Practice Parameters, but the current TF did not regard a single hypnotics specifically among demented older adults (a popu- case report as sufficient evidence for a recommendation. lation of interest for this review, see section 5.4 in this paper) There is insufficient evidence to support the use of pre- are scarce,63 but their cognitive and other vulnerabilities would scribed sleep-wake scheduling as a treatment for patients appear to place them at even greater risk than non-demented with ASWPD (versus no treatment). No recommendation. elderly adults. Benzodiazepines in particular are associated with an in- 5.1.2 Timed Physical Activity/Exercise for Patients with creased incidence of confusion, impaired motor performance, ASWPD anterograde amnesia, daytime sleepiness, and physiologic No recommendation was made in the 2007 Practice Param- dependence.63 The newer generation nonbenzodiazepine ben- eters, and no new studies were identified. zodiazepine receptor agonists (e.g., zolpidem, zaleplon, eszop- There is no evidence to support the use of timed physi- iclone) have shorter half-lives and fewer overall side effects, cal activity or exercise as a treatment for patients with but high quality data to support their use with demented older ASWPD. No recommendation. adults are nonexistent.63,64 Commonly used over-the-counter antihistamines have very high rates of side effects, includ- 5.1.3 Strategic Avoidance of Light for Patients with ASWPD ing cognitive impairment, daytime somnolence, and anticho- No recommendation was made in the 2007 Practice Param- linergic responses.63 Trazodone, a sedating antidepressant, is eters, and no new studies were identified. widely used off-label as a hypnotic, despite the fact that there There is no evidence to support the use of strategic avoid- is virtually no evidence-based data to support its efficacy ance of light as a treatment for patients with ASWPD. No with older adults.65 Moreover, it is associated with significant recommendation. risks, including priapism, orthostatic hypotension, and car- diac arrhythmias.65 Finally, the off-label use of neuroleptics 5.1.4 Light Therapy for Patients with ASWPD for dementia-related behavioral disturbances (including sleep No treatment trials of light therapy in ASWPD have been disturbances) is associated with a “black box” warning, due to published since the 2007 Practice Parameters, which recom- increased mortality risks (approximately twofold higher than mended this therapy as an OPTION. Most of the previously that associated with placebo-treated patients), mostly due to referenced studies68–71 were not included for the current cardiovascular or infectious causes.66 analysis, as participants were not clearly diagnosed with ASWPD. Discrete benefits from this treatment are difficult 5.0 RECOMMENDATIONS FOR TREATMENTS OF to ascertain given methodological limitations within the 72,73 INTRINSIC CRSWDs two reviewed studies (discussed further below), and the cumulative level of evidence was VERY LOW (Appendix, There were 8 different treatments, in addition to combi- Table S1). nation treatments, that were reviewed for the four intrinsic Only one randomized Advanced Sleep Phase Disorder CRSWDs included in this guideline. The strengths of the rec- (ASPD)/ASWPD treatment trial was identified (also refer- ommendations, for those intrinsic CRSWD treatments that enced in the previously published Practice Parameters).72 A have recommendations, are shown in Table 5. The full recom- parallel group design was used to test 28 days of daily evening mendation statements, including all GRADE components that exposure to a white broad-spectrum light (~265 lux) versus a were considered when determining the direction and strength dim red light control (~2 lux) among 47 patients (mean age of the recommendation, are shown in Table 6. 70.0 ± 6.4 years). The light source consisted of a rice paper shade over a vertical light tube, and the protocol instructed 5.1 Recommendations for the Treatment of ASWPD subjects to sit within 1 meter of the light source, for 2–3 hours ASWPD is characterized by a stable advance (earlier tim- in duration, ending 1 hour before habitual bedtime. There ing) of the major sleep episode, such that habitual sleep onset were no significant posttreatment group differences with

1207 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

Table 5—Overview of AASM recommendation status for Intrinsic CRSWD treatments Treatment ASWPD DSWPD N24SWD ISWRD Prescribed sleep-wake No Recommendation No Recommendation No Recommendation No Recommendation scheduling Timed physical activity/ No Recommendation No Recommendation No Recommendation No Recommendation exercise Strategic avoidance of light No Recommendation No Recommendation No Recommendation No Recommendation 5.4.4a WEAK FOR Light therapy 5.1.4a WEAK FOR (adults) No Recommendation No Recommendation (elderly with dementia) Sleep-promoting 5.4.5a STRONG AGAINST No Recommendation No Recommendation No Recommendation medications (elderly with dementia) 5.2.6.1a WEAK FOR (adults with and without depression) 5.4.6.1a WEAK AGAINST 5.3.6.1a WEAK FOR (elderly with dementia) 5.2.6.2.1a WEAK FOR (blind adults) Timed oral administration of No Recommendation (children/adolescents melatonin or agonists 5.4.6.2a WEAK FOR without comorbidities) No Recommendation (children/adolescents with (sighted) neurologic disorders) 5.2.6.2.2a WEAK FOR (children/adolescents with psychiatric comorbidities) Wakefulness-promoting No Recommendation No Recommendation No Recommendation No Recommendation medications Other somatic interventions No Recommendation No Recommendation No Recommendation No Recommendation No Recommendation (adults) 5.4.9.1a WEAK AGAINST (combination treatment Combination treatments No Recommendation 5.2.9.2a WEAK FOR (light No Recommendation of light and melatonin for therapy + multicomponent demented, elderly patients) behavioral interventions for children/adolescents)

respect to most predefined CRITICAL outcomes, namely years). The light source consisted of two light boxes (proxim- circadian phase (urinary 6-sulfatoxymelatonin [aMT6s] acro- ity to source not specified), and therapy was for 2 hours in phase), ISL (actigraphy and subjective), SOT (actigraphy and duration, between 20:00 and 23:00 (ending before habitual subjective), SOffT (actigraphy and subjective), and TST (sub- bedtime). The treatment significantly delayed circadian phase jective). Total sleep time (actigraphy) significantly decreased (CBTMin, mean difference 141.00 minutes [CI 26.10–255.90]), posttreatment within the active treatment group (mean differ- and increased TST (PSG, mean difference 51.30 minutes [CI ence 34.62 minutes [CI −0.96 to −68.28]), possibly due to a 2.69–99.91]), but both values crossed the threshold of the pre- nonsignificant delay in SOT. Importantly, while photosensors determined clinically significant minimal change (see Table 2 attached to the light source indicated good compliance with and Appendix, Table S1). There were no significant changes the scheduled light therapy on and off times, light exposure in ISL, SOT, or SOffT (PSG) posttreatment relative to the con- data on wrist worn photometers suggested that, on average, trol condition (Appendix, Table S1). patients were only adjacent to the light source for about half The results of studies that tested evening light therapy in of the scheduled treatment duration. Nonetheless, the treat- patients complaining of insomnia (not eligible for the current ment was well tolerated, and the majority of the patients who review, as subjects were not discreetly diagnosed with AS- received the white broad spectrum light source asked to keep WPD) are generally favorable. For example, in studies of pa- it for long-term personal use. tients with early-morning awakenings, evening light delayed Given the low intensity of light used in this study, the TF circadian timing,70,71 delayed SOffT71 and increased TST,70,71 questioned the applicability of the results, and elected to ex- although positive effects were not always observed.69 Eve- pand the GRADE analysis to include a non-randomized trial ning light therapy also phase delayed circadian timing and (Appendix, Table S1). A parallel group design study by SOT in patients with sleep maintenance insomnia.68 In other- Campbell and colleagues73 (also referenced in the previous wise healthy older adults with sleep complaints, evening light Practice Parameters) tested 12 days of a daily evening expo- therapy delayed the DLMO, but there was no observed effect sure to bright white light (4,000 lux) versus a dim red light (~50 on ISL and TST, while sleep timing remained fixed by the lux) control in 16 patients with ASWPD (mean age 70.4 ± 4.9 study protocol.74

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1208 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

Table 6—Recommendation statements for treatment of patients with intrinsic CRSWDs. Direction and Benefits/ Treatment Strength of Quality of Harms Patients’ Values and (PICO Question) Recommendation Statement Recommendation Evidence Assessment Preferences Advanced Sleep-Wake Phase Disorder (ASWPD) 5.1.4 Light 5.1.4a The TF suggests that clinicians treat WEAK FOR VERY LOW Benefits The majority of patients therapy adult ASWPD patients with evening light closely would use this treatment. (PICO Question 4) therapy (versus no treatment) balanced with harms Delayed Sleep-Wake Phase Disorder (DSWPD) 5.2.6 Timed oral 5.2.6.1a The TF suggests that clinicians WEAK FOR LOW Uncertainty in The majority of patients administration treat DSWPD in adults with and without the estimates would use this treatment. of melatonin or depression with strategically timed melatonin of benefits/ agonists (versus no treatment) harms (PICO Question 6) 5.2.6.2.1a The TF suggests that clinicians WEAK FOR MODERATE Uncertainty in The majority of patients treat children and adolescents with DSWPD the estimates would use this treatment, (and no comorbidities) with strategically timed of benefits/ with appropriate informed melatonin (versus no treatment) harms consent from the patient and caregiver. 5.2.6.2.2a The TF suggests that clinicians WEAK FOR LOW Uncertainty in The majority of patients treat children and adolescents with DSWPD the estimates would use this treatment, comorbid with psychiatric conditions with of benefits/ with appropriate informed strategically timed melatonin (versus no harms consent from the patient treatment) and caregiver. 5.2.9 Combination 5.2.9.2a The TF suggests that clinicians treat WEAK FOR LOW Benefits The majority of patients Treatments children/adolescents with DSWPD with post- outweigh would use this treatment, awakening light therapy in conjunction with harms particularly with active behavioral treatments (versus no treatment) caregiver support. Non-24-Hour Sleep-Wake Rhythm Disorder (N24SWD) 5.3.6 Timed oral 5.3.6.1a The TF suggests that clinicians WEAK FOR LOW Benefits The majority of patients administration use strategically timed melatonin for the outweigh would use this treatment. of melatonin or treatment of N24SWD in blind adults (versus harms agonists no treatment) (PICO Question 6) Irregular Sleep-Wake Rhythm Disorder (ISWRD) 5.4.4 Light 5.4.4a The TF suggests that clinicians treat WEAK FOR VERY LOW Benefits The majority of well- Therapy ISWRD in elderly patients with dementia with closely informed patients and/or (PICO Question 4) light therapy (versus no treatment) balanced with caregivers of would elect to harms use this treatment. 5.4.5 Sleep- 5.4.5a The TF recommends that clinicians STRONG NONE* Harms The vast majority of well- promoting avoid the use of sleep-promoting medications AGAINST outweigh informed patients and/or medications to treat demented elderly patients with benefits caregivers would NOT elect (PICO Question 5) ISWRD (versus no treatment) to use this treatment. 5.4.6 Timed oral 5.4.6.1a The TF suggests that clinicians WEAK AGAINST LOW Harms The majority of patients administration avoid the use of melatonin as a treatment for outweigh and/or caregivers would of melatonin or ISWRD in older people with dementia (versus benefits NOT elect to use this agonists no treatment) treatment. (PICO Question 6) 5.4.6.2a The TF suggests that clinicians use WEAK FOR MODERATE Benefits The majority of patients strategically timed melatonin as a treatment outweigh and/or caregivers would for ISWRD in children/adolescents with harms elect to use this treatment. neurologic disorders (versus no treatment) 5.4.9 Combination 5.4.9.1a The TF suggests that clinicians WEAK AGAINST VERY LOW Harms The majority of patients treatments avoid the use of light therapy combined with outweigh and/or caregivers would melatonin in demented, elderly patients with benefits NOT elect to use this ISWRD (versus no treatment) treatment.

*Although no randomized controlled trials have examined sleep-promoting medications for the treatment of ISWRD, other extant literature indicates that administration of hypnotics to demented elderly patients increases risks of falls and other untoward outcomes (see separate “Harms and Adverse Effects” section).

1209 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

5.1.4a The TF suggests that clinicians treat adult ASWPD 5.2 Recommendations for the Treatment of DSWPD patients with evening light therapy (versus no treatment). DSWPD is characterized by habitual sleep-wake timing that [WEAK FOR] is delayed, usually greater than two hours, relative to conven- Summary: No treatment trials of light therapy in ASWPD tional or socially acceptable timing. Affected individuals com- have been published since the 2007 Practice Parameters, which plain of difficulty falling asleep at a time required to obtain recommended this therapy as an OPTION. The largest effects sufficient sleep duration on a school or work night, and experi- were seen after a 12-day treatment of 2 hours of bright white ence difficulties arising at the required times. When allowed to broad spectrum light (~4,000 lux) from 2 light boxes (proxim- follow his or her preferred schedule, sleep quality and quantity ity to source not specified), timed to occur daily between 20:00 are typically reported as normal.5 and 23:00, and ending before habitual bedtime.73 Nevertheless, the overall quality of evidence derived from the analyses of 5.2.1 Prescribed Sleep-Wake Scheduling for Patients with two publications72,73 was VERY LOW (Appendix, Table S1), DSWPD with potential benefits of light therapy closely balanced with The previous recommendation was designated as an OP- the harm/burden. Associated risks are minimal, as detailed TION (specifically with respect to chronotherapy, a prescribed separately in the “Harms and Adverse Effects” section. Pa- progressive delay of the sleep/wake schedule until the desired tients report reasonable compliance and high satisfaction with schedule is reached), based upon two studies with adult partici- this treatment,72 and light boxes are available over the counter pants.75,76 The 1993 study by Ito and colleagues,75 reviewed pre- in the U.S., at relatively affordable prices. Thus, the majority viously, was not included in the current analysis, as it did not of well-informed patients would choose light therapy versus investigate discrete sleep outcomes, but instead incorporated no treatment. subjective assessments of global functioning. The 1981 study by Czeisler and colleagues76 was excluded due to a low number 5.1.5 Sleep-Promoting Medications for Patients with ASWPD of subjects (n = 5). There is one report of an adult DSWPD No recommendation was made in the 2007 Practice Param- patient who developed free-running circadian rhythms after eters, and no new studies were identified. engaging in this treatment.77 There is no evidence to support the use of sleep-promot- Although ineligible for current analyses, 3 studies were ing medications as a treatment for patients with ASWPD. published subsequent to availability of the previous Practice No recommendation. Parameters that may bear relevance to the use of prescribed sleep-wake scheduling (other than chronotherapy) as a ther- 5.1.6 Timed Oral Administration of Melatonin or Agonists for apy for adolescent patients.78–80 Two groups79,80 described Patients with ASWPD potentially meaningful outcomes, but the parameters of inter- The administration of a low dose of melatonin after early est were different from those specifically defined by the TF, morning awakenings and upon final arising in the morning and, in the case of the de Sousa study,80 all participants were was indicated as an OPTION for ASWPD in the 2007 Practice healthy adolescents (i.e., not afflicted with DSWPD). Similarly, Parameters, based on expert consensus alone. No new studies the Sharkey group (2011) studied a cohort with subthreshold were identified. DSWPD (the general CRSWD social/occupational impairment There is no evidence to support the use of melatonin criterion was not applied), such that participants did not meet or agonists as a treatment for patients with ASWPD. No inclusion criteria for this review.78 While a 6-day prescribed recommendation. advanced sleep schedule (with adjunctive strategic avoidance of evening light) resulted in concomitant advances in DLMO, 5.1.7 Wakefulness-Promoting Medications for Patients with the majority of participants exhibited decreased TST. ASWPD There is insufficient evidence to support prescribed sleep- No recommendation was made in the 2007 Practice Param- wake scheduling as a stand-alone treatment (versus no eters, and no new studies were identified. treatment) for patients with DSWPD. No recommendation. There is no evidence to support the use of wakefulness- promoting medications as a treatment for patients with 5.2.2 Timed Physical Activity/Exercise for Patients with ASWPD. No recommendation. DSWPD No recommendation was made in the 2007 Practice Param- 5.1.8 Other Somatic Interventions for Patients with ASWPD eters, and no new studies were identified. No recommendation was made in the 2007 Practice Param- There is no evidence to support the use of timed physi- eters, and no new studies were identified. cal activity or exercise as a treatment for patients with There is no evidence to support the use of other somatic DSWPD. No recommendation. interventions as a treatment for patients with ASWPD. No recommendation. 5.2.3 Strategic Avoidance of Light for Patients with DSWPD No recommendation was made in the 2007 Practice Parameters. 5.1.9 Combination Treatments for Patients with ASWPD A relevant open-label study was published subsequently, but No recommendation was made in the 2007 Practice Param- it did not meet inclusion criteria for the present analysis, as pre- eters, and no new studies were identified. defined critical outcomes were not employed (i.e., solely submea- There is no evidence to support the use of combination sures from the Pittsburgh Sleep Quality Index were utilized).81 treatments for patients with ASWPD. No recommendation. Adult subjects with DSWPD + ADHD were instructed to wear

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1210 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders amber glasses that blocked wavelengths ≤ 530 nm “…from sun- Two of these publications91,92 did not meet inclusion criteria down until bedtime every evening,” for a minimum of 3 hours, for the present review, due to an insufficient number of sub- and for a period of 2 weeks. Concomitant instructions included jects,91 or due to design and reporting limitations92 that hin- the use of only floor and table lamps (i.e., avoidance of overhead dered comparisons with the included investigations. One study lights) during the evening. If a participant awoke during the was published subsequent to availability of the previous Prac- night, he/she was instructed to don the glasses prior to light tice Parameters.93 The three reviewed investigations provide exposure. In addition, subjects were given specific instructions contradictory information regarding sleep/circadian-related to avoid/minimize caffeine, nicotine, and alcohol. Outcomes effects of melatonin among adults with DSWPD. Level of re- were compared to a weeklong baseline assessment period. As viewed evidence: LOW (Appendix, Table S2). determined by the Pittsburgh Sleep Quality Index, significant In a randomized double-blinded placebo-controlled study improvements in TST, ISL, and sleep quality were noted. In a (parallel design, n = 11, mean age 28.2 ± 5.7 years27), Mundey separate study potentially related to the treatment of DSWPD, and colleagues utilized either 0.3 or 3.0 mg melatonin (versus adult insomnia patients who wore “blue blocker” (< 550 nm) placebo) between 15:00–21:30 (1.5–6.5 hours before baseline glasses during the 3 hours prior to habitual bedtime demon- DLMO), with an advance in timing of 1 hour after 2 weeks, for strated improved subjective sleep quality82 compared with the a total treatment duration of 29 days. No improvements in ac- placebo intervention (yellow-tinted glasses that blocked only tigraphically determined sleep parameters were observed, and ultraviolet light). Importantly, there are no tangible risks asso- our analysis demonstrated no group difference with respect ciated with these interventions. to the timing of DLMO27 (Appendix, Table S2). As the pres- There is insufficient evidence to support the use of stra- ent review did not analyze outcomes relative to the timing of tegic avoidance of light as a treatment for patients with melatonin administration, however, it is important to note that DSWPD (versus no treatment). No recommendation. the authors reported an inverse relationship between the timing of melatonin administration (irrespective of dose) and the 5.2.4 Light Therapy for Patients with DSWPD magnitude of DLMO phase advance, such that earlier timing The previously published recommendation was designated of the former (in relation to DLMO) resulted in greater phase as a GUIDELINE and was based on two studies.28,83 The advances. No such correlation was identified with respect to

Rosenthal study was not analyzed for current purposes due CBTMin (assessed only within the active treatment groups). to a lack of rigorously reported sleep-related outcomes.83 The The Kayumov and Rahman studies (same investigative Cole study28 is included in a separate section (see Combination group) demonstrated substantial select PSG-measured ben- Treatments, below). A separate open-label light therapy trial efits in sleep parameters (TST, ISL), but an affiliated circadian was identified,84 but ultimately rejected due to a small number phase marker was not employed.90,93 Solely the Mundey group of subjects (n = 6). Only one study pertaining to adult DSWPD included such a measure. The Rahman study93 (n = 20, ran- populations has been published subsequent to the release of the domized, double-blind, placebo-controlled, crossover design, previous CRSWD Practice Parameters in 2007 (also reviewed mean age 30.8 ± 12.4 and 35.6 ± 14.0 years for females and in the Combination Treatments section below).85 males, respectively) utilized 5 mg melatonin administered There is insufficient evidence to support efficacy of post- between 19:00–21:00, for a period of 28 days. The Kayumov awakening light therapy (monotherapy) as a treatment for study90 (n = 20, same design/age distribution) used the same DSWPD (versus no treatment). No recommendation. dose, but scheduled it at 19:00 the first week, between 19:00– 21:00 during the second and third weeks (according to sub- 5.2.5 Sleep-Promoting Medications for Patients with DSWPD jects’ preferences), and at a consistent time during the fourth No recommendation was made in the 2007 Practice week (average chosen time 21:00). Analyses among a subset of Parameters. depressed patients from both studies (n = 28) demonstrated a There are isolated reports regarding the use of hypnotics statistically significant increase in TST (mean difference 41.44 in DSWPD (typically as an adjunctive treatment with chrono- minutes, [CI 13.19–69.70]), but the level of evidence was down- therapy), but there is insufficient rigor in methodology for pur- graded due to inconsistency and imprecision (see Figure 2 and poses of evidence analysis.75,86 Two reports describe DSWPD Tables 2 and 3). More definitive results were obtained from an patients’ resistance to the effects of traditional hypnotics.87,88 analysis of a subset of non-depressed patients from the Rah- Nevertheless, a laboratory-based study that imposed a 4-hour man study (Appendix, Table S2) (n = 12, TST = 56.00 minutes phase advance on healthy subjects described sleep-related ben- [CI 48.51–63.49]). efits (PSG and subjective measures) with zolpidem.89 ISL was polysomnographically assessed with the same There is insufficient evidence to support the use of sleep- subcategorization of groups. Among the depressed subgroup promoting medications as a treatment for patients with (n = 28), sleep latency decreased by 43.52 minutes [CI −34.45 DSWPD (versus no treatment). No recommendation. to −52.60] (see Figure 3 and Appendix, Table S2). Among the non-depressed subjects from the Rahman study (n = 12), 5.2.6 Timed Oral Administration of Melatonin or Agonists for sleep latency decreased by 37.70 minutes [CI −31.75 to −43.65] Patients with DSWPD (Appendix, Table S2).

5.2.6.1 Melatonin for adult patients with DSWPD 5.2.6.1a The TF suggests that clinicians treat DSWPD The previously published recommendation was designated in adults with and without depression with strategically as a GUIDELINE, and was supported by four studies.27,90–92 timed melatonin (versus no treatment). [WEAK FOR]

1211 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

Figure 2—Meta-analysis of data for PSG determined TST in response to melatonin treatment of adult patients with DSWPD and comorbid depression.

Figure 3—Meta-analysis of data for PSG determined ISL in response to melatonin treatment of adult patients with DSWPD and comorbid depression.

Summary: The previously published recommendation was The data of 64 participants were utilized for actigraphy/ designated as a GUIDELINE. The overall quality of evidence sleep-related analyses. With respect to CRITICAL outcomes, from the analyses of the three accepted/reviewed studies27,90,93 sleep onset time favorably advanced in comparison to placebo was LOW (Figures 2 and 3 and Appendix, Table S2), and among the 0.1 and 0.15 mg/kg groups (mean difference −33.70 data regarding the sleep/circadian-related effects of melatonin minutes [CI −10.95 to −56.46] and −42.77 minutes [CI −21.77 were contradictory. Positive results were obtained with a 5 mg to −63.78], respectively), but the confidence interval of the dose timed between 19:00–21:00 (no circadian-based timing), value associated with the lower dose crossed the threshold of for a period of 28 days.90,93 The Rahman study93 was the sole the predetermined clinically significant minimal change (see study identified subsequent to publication of the previous Prac- Table 2 and Appendix, Table S3). Statistical differences were tice Parameters. Taking into account the discussion regarding not observed with the 0.05 mg/kg group. Nevertheless, sleep potential safety/adverse effects of melatonin (see separate latency improved among all three groups (statistical and clini- “Harms and Adverse Effects” section), the benefits/harms ratio cally significant differences) in comparison to placebo (mean remains uncertain, but clinical experience suggests frequent difference −38.39 minutes [CI −18.24 to −58.53], −44.24 min- acceptance of this treatment among adults versus no treatment. utes [CI −24.04 to −64.44], and −43.80 minutes [CI −24.06 to −63.54], respectively, in order of ascending dosage). 5.2.6.2. Melatonin for children/adolescents with DSWPD With respect to DLMO analyses, no significant differences were noted among any of the three treatment groups in com- 5.2.6.2.1 Melatonin for children/adolescents with DSWPD and no parison to placebo. However, the authors separately calculated comorbidities individualized outcomes based upon melatonin use in relation No studies were previously reviewed which directly ad- to both DLMO (circadian) and clock time of administration dressed the pediatric/adolescent population. Strategically (TOA). These analyses were not compatible with the Review timed melatonin at the dose specified below is effective for Manager software used in this project. Data were depicted children/adolescents with DSWPD. Level of reviewed evi- solely within figures (i.e., no raw representation). As with the dence: MODERATE (Appendix, Table S3). previously discussed Mundey study,27 a positive relationship One randomized, placebo-controlled double-blinded study between DLMO phase advances and an earlier circadian TOA was reviewed (unpublished raw data provided courtesy of au- was described (no relationship observed with respect to clock thor).29 Participants ranged in age from 6–12 years. The three TOA), but no differences were observed between the various active treatment groups received melatonin at dosages of 0.05 melatonin dosage groups. On the contrary, no advantages be- mg/kg, 0.1 mg/kg, and 0.15 mg/kg, with respective mean doses tween clock versus circadian TOA were demonstrated in rela- of 1.6 ± 0.4 mg (n = 16), 2.9 ± 0.9 mg (n = 19), and 4.4 ± 1.0 mg tion to sleep onset and initial sleep latency times. (n = 18). Seventeen participants were allocated to the placebo group. The duration of treatment was 6 nights, with instruc- 5.2.6.2.1a The TF suggests that clinicians treat children tions to take melatonin 1.5–2.0 hours prior to habitual bedtime and adolescents with DSWPD (and no comorbidities) (unclear if equivalent to habitual sleep onset time), with con- with strategically timed melatonin (versus no treatment). sistent nightly timing. [WEAK FOR]

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1212 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

Figure 4—Meta-analysis of data for DLMO in response to melatonin treatment of children/adolescents with DSWPD and comorbid psychiatric conditions.

Figure 5—Meta-analysis of data for actigraphically-determined SOT in response to melatonin treatment of children/ adolescents with DSWPD and comorbid psychiatric conditions.

Summary: This is a new recommendation in comparison to (mean difference −54.22 minutes [CI −31.67 to −76.78]) (see the prior Practice Parameter, as no studies were previously re- Figure 4). Actigraphically assessed sleep-onset time (n = 130) viewed which directly addressed the pediatric/adolescent pop- also advanced (mean difference −36.57 minutes [CI −16.96 to ulation. The level of reviewed evidence from a singular study29 −56.18]) (see Figure 5). Other actigraphically derived sleep pa- was MODERATE (Appendix, Table S3). Optimal results rameters were obtained only in the more recent study (n = 105). were obtained with a dose of 0.15 mg/kg, taken 1.5–2.0 hours A significant decrease in ISL was detected (mean difference prior to habitual bedtime, for 6 nights. Although no serious −18.70 minutes [CI −7.01 to −30.39], but the confidence interval adverse reactions have been described in relation to melatonin crossed the threshold of the predetermined clinically signifi- use to date, relevant concerns have been raised by select stud- cant minimal change (see Table 2 and Appendix, Table S4). ies with respect to the pediatric/adolescent population,94 and No significant group differences were noted with respect to rigorous long-term data are lacking (see separate “Harms and TST.95 Previously published subjective assessments failed to Adverse Effects” section). As such, the benefits/harms assess- demonstrate significant group differences in TST (n = 31), ISL ment is uncertain. Clinical experience nevertheless supports (n = 33), or sleep-onset (n = 33) or offset (n = 32) times.96 frequent acceptance of this therapy versus no treatment, with Although not accepted for analysis for this review due to con- appropriate informed consent from the patient and caregiver. cerns regarding patient heterogeneity, a randomized placebo- controlled study by Wasdell and colleagues (which contained 5.2.6.2.2 Melatonin for children/adolescents with DSWPD and psychi- an admixture of DSWPD and insomnia patients) explored the atric comorbidities effects of 5 mg controlled-release melatonin (10 days) among This is a new recommendation in comparison to the previ- approximately 50 children with neurodevelopmental disabili- ous Practice Parameters, as no studies specifically addressed ties (ages 2–18, mean 7.4 years).53 Melatonin was administered this patient population. Strategically timed fast-release mela- 20–30 minutes prior to the caregiver determined desired time tonin at dosages ranging from 3–5 mg may be effective for of sleep onset. Significant improvements were observed in sub- children/adolescents with DSWPD and psychiatric comorbidi- jectively and actigraphically recorded TST, ISL, and clinician- ties. Level of reviewed evidence: LOW (Appendix, Table S4) and caregiver-assessed overall sleep disorder severity and Two randomized, placebo-controlled studies by the same other functional and health dimensions. During the 3-month group examined the use of melatonin for DSWPD among chil- open-label phase of the trial, escalation of the melatonin dos- dren/adolescents with various psychiatric comorbidities (all age provided no definitive benefits. A separate follow-up open- were diagnosed with ADHD).95,96 Participants aged 6–12 years label prospective study (duration up to 3.8 years) demonstrated received fast-release melatonin (one of several instances where continued caregiver-reported benefits on sleep, overall health, melatonin formulation was specified) for 4 weeks at dosages behavior, education, and learning.54 of 3 or 5 mg, at either 18:00 or 19:00. The more recent study95 based dosage on weight (3 mg if < 40 kg; 5 mg if > 40 kg, 5.2.6.2.2a The TF suggests that clinicians treat children taken at 19:00), while the earlier protocol96 uniformly provided and adolescents with DSWPD comorbid with psychiatric 5 mg at 18:00. Combined analyses (n = 132) revealed an ad- conditions with strategically timed melatonin (versus no vance in DLMO of nearly 1 hour in comparison to placebo treatment). [WEAK FOR]

1213 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

Summary: This is a new recommendation in comparison to employed multicomponent treatment strategies, but failed to the previous Practice Parameters, as no studies specifically ad- mention or emphasize their nature in the published method- dressed this patient population. The overall quality of evidence ologies or to include proper controls for all components of the from the analyses of the two reviewed studies95,96 was LOW treatment. (see Figures 4 and 5 and Appendix, Table S4). A fast-release formulation of melatonin was utilized, with dosages ranging 5.2.9.1. Light/combination treatments for adults with from 3–5 mg, taken between 18:00–19:00 (no circadian-based DSWPD timing), for 4 weeks. In the pooled analysis, actigraphically There is no evidence to support efficacy of light therapy assessed sleep onset time advanced in conjunction with an (provided by means other than a “light box”) in association advance in the circadian phase marker (DLMO). Although no with concomitant behavioral instructions among adults. Level serious adverse reactions have been described in relation to of reviewed evidence: VERY LOW (Appendix, Table S6). melatonin use to date, relevant concerns have been raised by In a randomized parallel clinical trial (n = 54) involving select studies with respect to the pediatric/adolescent popu- adults (mean age 25 ± 6 years) with DSWPD, Cole and col- lation, and rigorous long-term data are lacking (see separate leagues (2002)28 used a light mask that exposed participants to “Harms and Adverse Effects” section). As such, the benefits/ 2700 lux of white broad spectrum light on closed eyelids (about harms assessment is uncertain. Clinical experience neverthe- 57 lux at the cornea) for a period of 26 days, and compared it less supports frequent acceptance of this therapy versus no to an inactive condition (0.1 lux of red light, estimated corneal treatment, with appropriate informed consent from the patient illuminance = 0.007 lux). Light therapy commenced (at an in- and caregiver. tensity of < 0.01 lux) 4 hours before the scheduled rise time, proceeded with ramped intensity for 1 hour, and remained at 5.2.7 Wakefulness-Promoting Medications for Patients with full brightness until arising. Volunteers were not informed of DSWPD the hypotheses of the study and were blinded to the treatment No recommendation was made in the 2007 Practice Param- received by the other group. The study personnel who inter- eters, and no new studies were identified. acted with participants and scored data were also blind to treat- There is no evidence to support the use of wakefulness- ment assignments. promoting medications as a treatment for patients with Both groups received concomitant behavioral instructions. DSWPD. No recommendation. Subsequent to the first 2 treatment days, subjects were asked to advance both bedtime and time of arising (in equal measure) on 5.2.8 Other Somatic Interventions for Patients with DSWPD a daily basis, to achieve a cumulative advance of at least 1 hour In the previous Practice Parameters, oral vitamin B12 ther- per week (≤ 15 minutes daily, unless he/she experienced spon- apy was described as “not indicated,” designated at the GUIDE- taneous awakening > 15 minutes ahead of schedule 2 days in a LINE level. No subsequent relevant studies (involving B12 or row). Subjects were also asked to stop advancing sleep if he/she separate somatic interventions) have been published. There achieved a preselected target wakeup time, but were allowed is no clear evidence of benefit from oral vitamin B12 among to change the target time if he/she wished. Participants were adults with DSWPD. Level of reviewed evidence: VERY LOW urged to avoid naps during the treatment interval, and were (Appendix, Table S5). provided specific instructions to minimize late-afternoon and In a multicenter study97 that contained predominantly adult evening light exposure on all treatment days. Specific interven- DSWPD patients (n = 50, mean age 26.8 ± 1.3 years), partic- tions included shading of windows between 17:00 and dusk and ipants received 3 mg total daily dose of vitamin B12 (three utilization of dark wraparound form-fitting goggles (2% trans- times daily divided dosing) versus placebo (double-blinded) mittance) if necessary to go outdoors during this time interval. for a period of 4 weeks. No clinically significant differences Subjects were also instructed to use the minimum practical were observed with respect to subjectively assessed TST and amount of artificial room light from 17:00 to bedtime. SOT (see Table 2 and Appendix, Table S5). Remaining sleep As determined by the acrophase of aMT6s excretion, there parameters were not compatible with the TF’s predefined clini- were no group phase differences. Similarly, there were no dif- cal outcomes (ISL was evaluated according to a 5-point Likert ferences in (predominantly actigraphically measured, n = 2 scale). sleep logs) sleep onset or offset times. When the group was There is insufficient evidence to support the use of oral subdivided according to whether the baseline acrophase was vitamin B12 (and no evidence to support alternate somatic later or earlier than 06:00 (n = 23 and n = 22, respectively, used interventions) among patients with DSWPD (versus no as a measure to divide groups by the degree to which they were treatment). No recommendation. physiologically delayed), significant differences were noted solely for the former group, but the confidence interval crossed 5.2.9 Combination Treatments for Patients with DSWPD the threshold of the predetermined clinically significant mini- Combination treatments for DSWPD were not addressed mal change (mean difference −83.00 minutes in comparison to separately in the previous Practice Parameters and associated the inactive condition [CI −8.09 to −157.91]) (see Table 2 and review papers, but one of the reviewed studies28 was addressed Appendix, Table S6). No significant differences were noted in the singular light therapy category. The combination treat- with respect to actigraphically determined sleep onset and off- ments sections throughout the paper contain studies that ex- set times among either subgroup. The light mask was never- plicitly denote dual treatment interventions, although the theless described as well tolerated, with little (actigraphically TF acknowledges that other reviewed studies may have also measured) sleep disturbance.

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1214 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

A separate group85 performed a randomized controlled trial children/adolescents) and light/behavioral multicomponent of 7 days duration (n = 18, mean age 28.2 ± 10.6 years) using interventions. post-awakening morning blue light in association with a grad- One non-blinded randomized controlled trial104 was identi- ually advancing sleep/wake schedule. The portable source con- fied (n = 40), composed of adolescents aged 13–18 (mean ages tained light-emitting diodes (LEDs; 470 nm peak wavelength 14.7 ± 1.7 and 14.7 ± 1.8 years in the active and control groups, with irradiance of 65 μW/cm2) attached to the lower rims of respectively). The minority of participants (20%) had mental spectacle frames, approximately 15 mm from the cornea, for health comorbidities, and was spread evenly across the two a duration of 2 hours, immediately after arising. While a bed- groups. Over a period of 8 weeks, active subjects (n = 23) were time was not prescribed, subjects were instructed to advance exposed to either post-awakening natural sunlight (when avail- wake times by 30 minutes daily. The presence or absence of able) or a broad spectrum lamp (~1000 lux, proximity to source blinding procedures was not specified. As compared to the not specified) for ≥ 30 minutes (2 hours maximum), with the control group (lighting exposure unspecified), DLMO ad- time of administration advanced 30 minutes daily from “natu- vanced by 2.53 hours (reported verbally and depicted in a fig- ral” wake time, until a target time of 06:00 was reached (3–5 ure). However, there were no group differences in subjectively weeks). Light therapy was subsequently discontinued and a reg- assessed TST, SOT, or SOffT. Among other design limitations, ular rise time between 06:30–07:30 was advised. Concomitant many details were missing with respect to methodology, and multicomponent behavioral education/interventions (includ- there was high imprecision in the reported results. ing instructions to “reduce evening light”) were provided in Despite the present uncertainty as to benefits, there are few six 45–60 minute sessions (with parental involvement), either major risks associated with a trial of one of these “non-light weekly (first 4 sessions) or biweekly (last 2 sessions). Compli- box” interventions. A review of adverse effects associated ance was monitored with sleep diaries, but affiliated data were with light interventions in general is addressed elsewhere (see not provided. An objective measure was not employed to mea- “Harms and Adverse Effects” section). It is interesting that the sure light therapy compliance specifically. The control group light mask worn during the sleep episode in the Cole protocol28 was designated to a waitlist (n = 17). was well tolerated, with little (actigraphically measured) sleep Solely completed participants’ data were analyzed (i.e. an disturbance, as compliance with post-awakening light therapy intention-to-treat analysis was not undertaken). With respect can be poor.98 to CRITICAL outcomes during weekday assessments, sig- A laboratory-based study from Figueiro and Rea shows nificant differences were detected with respect to subjective promise for future “light masks” in the treatment of CRSWDs, TST (mean difference 72.00 minutes [CI 37.35 to 106.65]), and based upon optimal eyelid transmittance using green LEDs ISL (mean difference −43.10 minutes [CI −22.46 to −63.74]). (max wavelength of 527 nm) and individualized dose assess- Sleep onset and offset times also demonstrated significant ments to predict optimal circadian benefit.99 Particularly rel- group differences, but the confidence interval crossed the evant to the discussion of combination therapies, at least three threshold of the predetermined clinically significant minimal laboratory-based studies among healthy adults describe a change (mean difference −42.00 minutes [CI −2.74 to −81.26] synergistic effect (with respect to circadian phase advances) and −23.00 minutes [CI −0.87 to −45.13], respectively) (see when strategically timed light and melatonin therapy are used Table 2 and Appendix, Table S7). Weekend assessments together.100–102 In addition, although physiologic phase assess- demonstrated beneficial significant differences for sleep onset ments were not employed, a separate case series described suc- times (mean difference −93.9 minutes [CI −49.09 to −138.71]). cess with this treatment combination among DSWPD patients Significant favorable differences were also noted with respect in the field (age range 15–60 years). Three milligrams melatonin to ISL and sleep offset times, but the confidence interval again was taken 2 hours prior to desired bedtime in conjunction with overlapped the threshold of the predetermined clinically sig- either outdoor or 5000 lux white broad spectrum light exposure, nificant minimal change (mean difference −26.5 minutes [CI for a minimum of 30 minutes, between 06:00–08:00. During a −4.37 to −48.63] and −51.0 minutes [CI −10.82 to −91.18], re- median follow-up period of 6.4 weeks, 82% reported improve- spectively). No weekend differences were noted for TST. With ments of sleep patterns, with a mean sleep phase advance of ap- the exception of weekday sleep offset times, 6-month follow- proximately 2 hours.103 Reported outcomes were unfortunately up assessments (n = 15) revealed no significant differences in not compatible with those selected by the TF for this review. comparison to the values observed immediately posttreatment. There is insufficient evidence to support the use of novel forms of light therapy (i.e., via means other than a “light 5.2.9.2a The TF suggests that clinicians treat children box”), in association with concomitant behavioral instruc- and adolescents with DSWPD with post-awakening light tions among adults with DSWPD (versus no treatment). No therapy in conjunction with behavioral treatments (versus recommendation. no treatment). [WEAK FOR] Summary: This is a new recommendation, based both upon 5.2.9.2 Light/combination treatments for children/ the novel cohort (solely children/adolescents) and light/behav- adolescents with DSWPD ioral multicomponent interventions. The level of reviewed evi- Post-awakening light therapy in conjunction with behav- dence104 was LOW (Appendix, Table S7), and solely weekday ioral instructions is effective among adolescents with DSWPD, data were considered with respect to determination of the rec- but a physiologic circadian correlate is lacking. Level of re- ommendation, as this information is presumably most relevant viewed evidence: LOW (Appendix, Table S7). This is a new in the clinical setting. Light therapy occurred via exposure to recommendation, based both upon the novel cohort (solely natural sunlight (when available), or with use of a white broad

1215 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al. spectrum lamp (~1000 lux, proximity to source not specified), blind individuals in the absence of circadian photoreception.111 for ≥ 0.5 hours (2 hours maximum), with the time of adminis- As with prescribed sleep-wake scheduling, however, there have tration advanced by 0.5 hours daily from “natural” wake time, not been any published trials. The evidence is therefore indirect until a target time of 06:00 was reached. Light therapy was sub- and insufficient to serve as the basis of a recommendation. sequently discontinued, and behavioral interventions ensued. There is no evidence to support the use of timed physi- Follow-up data are promising. Overall, a benefits/harms ratio cal activity or exercise in patients with N24SWD. No analysis favors a trial of treatment, as children/adolescents with recommendation. DSWPD represent a particularly challenging patient population (for a multitude of reasons), and the suggested interventions 5.3.3 Strategic Avoidance of Light for Patients with N24SWD pose no apparent safety concerns (see separate “Harms and Ad- There is one case report that includes strategic avoidance of verse Effects” section). Clinical experience suggests that mo- light, and it is included below in the combination treatments tivated patients would accept this treatment option versus no section.112 The previous Practice Parameters1 provided no per- treatment, particularly with active caregiver support. tinent recommendations. There is no evidence to support the use of strategic avoid- 5.3 Recommendations for the Treatment of N24SWD ance of light (as monotherapy) in patients with N24SWD. N24SWD occurs when the hypothalamic circadian pace- No recommendation. maker fails to entrain (synchronize) to the 24-hour day. As a result, individuals can suffer from periodic nighttime insom- 5.3.4 Light Therapy for Patients with N24SWD nia and daytime somnolence as the circadian rhythms of sleep The previous Practice Parameters recommended this treat- propensity and alertness drift in and out of synchrony with the ment at an OPTION level (for sighted individuals), based on 24-hour day.5 The condition primarily occurs in blind individu- 5 case series/reports,112–116 all composed of 1–2 subjects. The als, and at least 50% of the totally blind (i.e., those with no light small numbers of study participants prevented inclusion for the perception) are thought to suffer from the disorder.105 While the present report, and no studies have been published subsequently. etiology in the blind is a loss of photic input to the pacemaker,106 There is some experimental evidence that light is capable of the pathophysiology among sighted individuals is unknown. resetting the circadian pacemaker in the absence of conscious As stated above, entrainment status was uniquely utilized as light perception,117 and most blind individuals who retain such an outcome measure, as it physiologically defines this disorder photic input to the circadian pacemaker would not be expected (just as elevated blood pressures characterize essential hyper- to have N24SWD. However, some such individuals may have tension). This was the sole outcome variable rated as CRITI- insufficient exposure to light to maintain entrainment and CAL for N24SWD. therefore timed light exposure may be a potential therapeutic intervention in a subset of blind individuals with N24SWD. 5.3.1 Prescribed Sleep-Wake Scheduling for Patients with However, as noted, there have been no studies that examined N24SWD this question. The previously published Practice Parameters1 recom- There is insufficient evidence to support the use of light mended prescribed sleep-wake scheduling at an OPTION therapy in patients with N24SWD (versus no treatment). level, noting that it may be a useful method to improve the No recommendation. entrainment of circadian rhythms among sighted patients with N24SWD. However, this guideline was created in the absence 5.3.5 Sleep-Promoting Medications for Patients with N24SWD of discrete evidence, as there have been no published trials The previous Practice Parameters did not make any rec- among sighted or blind patients. ommendation for the use of sleep-promoting medications in There is some evidence that sleep timing (independent of N24SWD, and no new studies were identified. the timing of light exposure) is able to reset the circadian pace- There is no evidence to support the use of sleep- maker in humans,20,107–109 and prescribed sleep-wake schedul- promoting medications in patients with N24SWD. No ing therefore represents a potential therapeutic intervention recommendation. for N24SWD. However, this evidence is indirect, as data were culled from basic science experiments conducted among non- 5.3.6 Timed Oral Administration of Melatonin or Agonists for clinical populations. Patients with N24SWD There is no evidence to support the use of prescribed sleep-wake scheduling in patients with N24SWD. No 5.3.6.1 Melatonin for blind adult patients with N24SWD recommendation. The previously published recommendation was designated at the GUIDELINE level,1 based upon 4 case reports118–121 and 5.3.2 Timed Physical Activity/Exercise for Patients with 5 small observational studies.122–126 Some of these studies, and N24SWD others, have examined important treatment parameters and There was no recommendation in the previous Practice Param- outcomes, including dose,125,127,128 phase angle of entrainment,126 eters1 regarding timed physical activity/exercise as a treatment and circadian time of administration,129 but many were small, for patients with N24SWD. Physical activity has been demon- often uncontrolled trials. Three placebo-controlled, crossover strated to reset the timing of the circadian pacemaker among studies from the previous Practice Parameters were eligible for healthy subjects110 and therefore represents another potential the current review,122–124 and, despite the relatively large odds therapy. Indeed, the circadian pacemaker can be entrained in ratio, the overall level of evidence for the CRITICAL outcome

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1216 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

Figure 6—Meta-analysis of evidence for entrainment as a result of melatonin treatment of blind adult patients with N24SWD.

(circadian entrainment) was LOW (Appendix, Table S8), as based upon 4 case series.115,130–132 None of these studies were the data are from small observational investigations. eligible for the current review, based upon insufficient num- Sack and colleagues (n = 7, mean age 47.3 ± 5.0 years) ad- bers of subjects. ministered 10 mg of melatonin 1 hour prior to preferred bed- There is insufficient evidence to support the use of mela- time,123 while Lockley et al. (n = 7, mean age 44.6 ± 8.4 years)122 tonin among sighted patients with N24SWD (versus no and Hack et al. (n = 10, mean age 48.2 ± 12.5 years)124 adminis- treatment). No recommendation. tered 5 mg and 0.5 mg of melatonin, respectively, at 21:00. The duration of treatment for these studies was 26–81 days. Re- 5.3.7 Wakefulness-Promoting Medications for Patients with sults (excluding subjects without placebo control or complete N24SWD CRITICAL outcome data) were combined for meta-analysis The previous Practice Parameters did not make any per- of entrainment as a dichotomous outcome (Yes/No), using the tinent recommendations regarding the use of wakefulness- Mantel-Haezel test (see Figure 6 and Appendix, Table S8). promoting medications in N24SWD, and no new studies were The odds ratio for entrainment was 21.18 [CI 3.22–39.17]. In identified. other words, the likelihood of entrainment with melatonin As noted above, both sleep and activity have the potential was ~21 times higher in comparison to placebo. Although the to reset the circadian pacemaker and therefore it is reasonable evidence level can only be graded as LOW due to the fact that to think that medications that promote either sleep or wakeful- these are small observational studies, this is the best evidence ness might be useful in the treatment of this disorder, either by to date that melatonin is an effective treatment for N24SWD. entrainment of the biological clock or by improving alertness Taken together, these placebo-controlled studies represent the and sleep when individuals are sleeping out of synchrony with largest number of N24SWD patients whose entrainment status the circadian pacemaker. Similar to sleep-wake scheduling was assessed subsequent to melatonin treatment. and physical activity however, there have not been any studies of either of these approaches in N24SWD. 5.3.6.1a The TF suggests that clinicians use strategically There is no evidence to support the use of wakefulness- timed melatonin for the treatment of N24SWD in blind promoting medications in patients with N24SWD. No adults (versus no treatment). [WEAK FOR] recommendation. Summary: This recommendation was designated at the GUIDELINE level (for the blind) in the previous Practice Pa- 5.3.8 Other Somatic Interventions for Patients with N24SWD rameters.1 Only 3 studies122–124 met inclusion criteria for the The previous Practice Parameters cited “insufficient evi- present analysis and the level of evidence from these small dence” to support the use of oral vitamin B12 for use in sighted trials was LOW (Figure 6 and Appendix, Table S8). Doses individuals with N24SWD (OPTION), based on 2 open-label ranged between 0.5–10.0 mg, and were administered either case reports.133,134 No studies have since been published per- 1 hour prior to preferred bedtime, or at a fixed clock hour taining to the use of vitamin B12 (or other alternate somatic (21:00), for a period of 26–81 days. Patient preference would interventions) for N24SWD. be expected to favor the use of easily obtained and inexpensive There is insufficient evidence to support the use of oral melatonin that requires once daily dosing. No serious adverse vitamin B12 (and no evidence to support alternate somatic reactions to melatonin have been described to date (see sepa- interventions) among patients with N24SWD (versus no rate “Harms and Adverse Effects” section), and therefore the treatment). No recommendation. benefits of use appear to outweigh any potential harms. A majority of well-informed patients and caregivers would therefore 5.3.9 Combination Treatments for Patients with N24SWD accept this treatment option versus no treatment. The previous Practice Parameters did not discuss the use of combination treatments in N24SWD. There are 4 uncontrolled 5.3.6.2 Melatonin for sighted patients with N24SWD case reports involving a combination of treatments in sighted The recommendation for sighted individuals was provided individuals with N24SWD.112,116,135,136 However, these studies at an OPTION level in the previous Practice Parameters,1 did not meet inclusion criteria for the present review.

1217 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

There is insufficient evidence to support the use of com- Both reviewed light therapy studies examined institutional- bination treatments in patients with N24SWD (versus no ized elderly subjects with dementia.137,146 In their randomized treatment). No recommendation. clinical trial of 50 nursing home patients (mean age 86 ± 8 years), Dowling and colleagues compared 60 minutes of light 5.4 Recommendations for the Treatment of ISWRD therapy scheduled from 09:30–10:30 (1.5 hours after the nurs- ISWRD is diagnosed when patients exhibit no clear cir- ing home’s set rise time) 5 days per week for a period of 10 cadian pattern of sleep-wake behavior. Afflicted individuals weeks, to a dim-light control condition consisting of indoor demonstrate wakefulness during conventional sleeping hours light of 150–200 lux.146 Bright light was delivered via white and bouts of sleep during the day.5 The condition is observed broad-spectrum light boxes, positioned on tables and located most commonly among patients with neurodevelopmental or approximately 30 to 34 inches from participants’ eyes. Light neurodegenerative disorders, and can pose particular chal- levels in the direction of gaze were monitored for each partici- lenges for caregivers. pant during each treatment session to maintain levels > 2500 lux. At the end of the intervention, average actigraphically 5.4.1 Prescribed Sleep-Wake Scheduling for Patients with measured TST did not differ between active and control groups. ISWRD A second open-label study tested light therapy of 2 hours du- There was no recommendation in the previous Practice ration between 09:00–11:00 among 14 patients (mean age 75 Parameters regarding prescribed sleep-wake scheduling as a years, range 61–83) with sleep disturbances. Patients were stand-alone treatment for patients with ISWRD, and no new treated for 4 weeks with white broad-spectrum tabletop light studies were identified. However, prescribed schedules have boxes that produced 3000–5000 lux and were positioned at a been tested in combination with other modalities (see section distance of 1 meter from the participants. As with the Dowl- 5.4.9, “Combination Treatments”). ing study, no improvements in TST (subjective) were demon- There is no evidence to support the use of prescribed strated, but the study did show improvement of caregiver-rated sleep-wake scheduling as a stand-alone treatment for pa- behavioral symptoms, including decreases in wandering, vio- tients with ISWRD. No recommendation. lent behavior, restlessness, and symptoms of delirium.137 Other research that was ineligible for this review may none- 5.4.2 Timed Physical Activity/Exercise for Patients with theless inform clinicians’ decision-making regarding the use ISWRD of light therapy among demented elderly patients with ISWRD. There was no recommendation in the previous Practice Pa- Riemersma van der Lek and colleagues147 randomized twelve rameters regarding timed physical activity/exercise as a sole assisted-living facilities in the Netherlands to common areas treatment for patients with ISWRD, and no new studies were that were lit with ceiling fixtures that emitted either bright identified. However, scheduled activity has been used in com- white broad spectrum (1000 lux) or dim light (300 lux). Par- bination with other modalities (see section 5.4.9, “Combination ticipants in the bright light facilities had lower TST compared Treatments”). to participants living in a dim light facility at two follow-up There is no evidence to support the use of timed physical assessments (6 weeks and 6 months after the change in facil- activity or exercise as a stand-alone treatment for patients ity light levels). Other outcomes, including cognitive deficits with ISWRD. No recommendation. and depression, were improved in association with the light intervention, however. Additional studies cited in the previous 5.4.3 Strategic Avoidance of Light for Patients with ISWRD Practice Parameters and associated review1,3 treated subjects There was no recommendation in the previous Practice Pa- with white broad-spectrum artificial light or outdoor light at rameters regarding strategic avoidance of light as a treatment levels ranging from 1000–8000 lux. Durations of daily light for patients with ISWRD, and no new studies were identified. exposure ranged from 45–120 minutes and treatment dura- There is no evidence to support the use of strategic avoid- tions lasted from 10 days to 11 weeks.138–145 Although presently ance of light as a treatment for patients with ISWRD. No defined CRITICAL sleep outcomes were not analyzed, some recommendation. of these studies showed positive effects of light exposure on 24-hour rest-activity rhythms, with more consolidated rest pe- 5.4.4 Light Therapy for ISWRD in Elderly Patients with riods at night and more activity and fewer naps during the day. Dementia In contrast to the present review, the 2007 Practice Param- 5.4.4a The TF suggests that clinicians treat ISWRD in eters1 did not separate ISWRD treatment recommendations by elderly patients with dementia with light therapy (versus patient subgroup. The section that follows (and other sections) no treatment). [WEAK FOR] is subcategorized as warranted. Summary: This recommendation was designated as an The recommendation in the 2007 Practice Parameters was OPTION in the 2007 Practice Parameters, and only one sub- designated as an OPTION, but only one137 of the previously sequent study has been published that met inclusion criteria reviewed studies138–145 and one new study146 met inclusion cri- for the current document.146 The cumulative level of reviewed teria for the current document. Although neither demonstrated evidence (2 studies)137,146 was VERY LOW (Appendix, Table improvements in CRITICAL outcomes, one137 showed that S9), and none of the TF-defined CRITICAL outcomes showed light therapy improved behavioral symptoms, an area of sig- improvement. Behavioral symptoms nevertheless improved in nificant clinical importance. The cumulative level of analyzed the sole study that measured this outcome.137 The interventions evidence was VERY LOW (Appendix, Table S9). consisted of white broad spectrum light therapy, 2500–5000

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1218 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

lux (~1 meter from participants), 1–2 hours in duration, be- of 6 mg slow-release melatonin versus placebo.151 Participants tween 09:00–11:00, for a period of 4–10 weeks.137,146 Benefits of were studied for a 2-week baseline period and were then ran- treatment are closely balanced with harm/burden. In addition domized to receive either 6 mg slow-release melatonin or pla- to the general side effects reported in the “Harms and Adverse cebo for 2 weeks at their usual bedtimes, followed by a 1-week Effects” section, other side effects in this population range from washout period and crossover to the second study period. complaints of eye irritation138 to agitation and confusion,141 and Mean TST estimated with actigraphy did not differ between these potential drawbacks should be considered when recom- the two groups. mending treatment. Furthermore, depending on the method Other research may inform clinical decision-making in this and setting of light delivery, treatment may be labor intensive, area. Singer and colleagues examined the effects of treatment and modest improvements in outcomes may not justify associ- with 2.5 mg slow-release or 10 mg immediate-release melatonin ated costs. Nevertheless, clinical experience suggests that the administered 1 hour before bedtime (versus placebo) on acti- majority of well-informed patients and/or caregivers of elderly, graphically estimated TST in patients with Alzheimer’s disease demented patients with ISWRD would choose light therapy in and sleep disturbance.152 The patients in this study did not have comparison to no intervention. discreet diagnoses of ISWRD and data were therefore not eligible for inclusion in the present review. Nevertheless, in keeping 5.4.5 Sleep-Promoting Medications for ISWRD in Elderly with the results of Serfaty and colleagues,151 this large, well- Patients with Dementia designed trial also failed to show an improvement in TST with There was no recommendation in the previous Practice Pa- either dose of melatonin compared to placebo. Riemersma van rameters regarding sleep-promoting medications for patients der Lek and colleagues147 published another potentially relevant with ISWRD, and no new studies have been published. Various combination study among dementia patients in assisted living in investigations have concluded, however, that hypnotic medica- whom ISWRD was not specifically identified. In contrast to the tions increase risks of adverse events within this population other studies, the melatonin-only arm (2.5 mg immediate-release (see “Harms and Adverse Effects” section). formulation administered approximately one hour before bedtime) demonstrated decreased actigraphic ISL and increased TST 5.4.5a The TF recommends that clinicians avoid the use compared to placebo. However, detrimental effects of melatonin of sleep-promoting medications to treat demented elderly on mood and daytime functioning were also observed. patients with ISWRD (versus no treatment). [STRONG AGAINST] 5.4.6.1a The TF suggests that clinicians avoid the use of Summary: This is a new recommendation in comparison melatonin as a treatment for ISWRD in older people with to the previous Practice Parameters, which did not address the dementia (versus no treatment). [WEAK AGAINST] use of sleep-promoting medications (other than melatonin) for Summary: Melatonin was deemed “not indicated” for the ISWRD. Although no relevant subsequent studies have been treatment of ISWRD in older people with dementia (OPTION) published, other extant literature indicates that administration in the previous Practice Parameters. The present recommen- of hypnotics to demented elderly patients increases risks of dation against melatonin treatment is based on one reviewed falls and other untoward outcomes Altered pharmacokinetics study that failed to show benefit with respect to the CRITICAL observed with aging may be one mechanism by which hypnot- outcome of TST (Level of evidence: LOW (Appendix, Table ics increase adverse events in older adults.148 Risk appears to S10).151 Furthermore, there is evidence that melatonin could be be increased even further in elderly patients with dementia,149 harmful in this population.147 Thus, the risk-benefit ratio sug- particularly when used in combination with other medica- gests that the potential for harms outweighs the possibility for tions150 (also see separate “Harms and Adverse Effects” sec- benefits. Clinical experience therefore dictates that the ma- tion). Thus, the risk of harm from use of hypnotics in demented jority of older patients with dementia and/or their caregivers elderly patients with ISWRD outweighs potential positive ef- would not favorably accept a trial of melatonin. fects. As such, the vast majority of well-informed patients and/ or caregivers would not select this treatment. 5.4.6.2 Melatonin in children/adolescents with ISWRD and neurologic disorders 5.4.6 Timed Oral Administration of Melatonin or Agonists for Melatonin was recommended at the OPTION level in the Patients with ISWRD previous Practice Parameters for children with various neurologic disorders,1 based upon four studies,153–156 none of which 5.4.6.1 Melatonin for elderly patients with dementia and was eligible for the current review, due to an insufficient num- ISWRD ber of participants or grouping of participants with different In the 2007 Practice Parameters, melatonin was deemed CRSWDs. One new eligible study was identified,157 and mela- “not indicated” for this specific population, based upon two tonin was shown to improve select predefined CRITICAL out- studies,151,152 only one of which was eligible for the current re- comes (TST, ISL), although the confidence interval associated view.151 No subsequently published relevant studies were iden- with both values crossed the threshold of the predetermined tified. Melatonin administration did not significantly improve clinically significant minimal change (seeTable 2). The level of the predefined CRITICAL outcome of TST. Level of evidence: reviewed evidence was MODERATE (Appendix, Table S11). LOW (Appendix, Table S10). Wright and colleagues157 performed a double-blind, ran- Twenty-five patients with dementia and ISWRD (mean age domized, controlled, crossover trial in 16 children (mean age 84.2 ± 7.6 years) were enrolled in a double-blind crossover trial 9 ± 2.9 years) with autism spectrum disorder. The protocol

1219 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al. consisted of 1-month baseline, 3-months melatonin vs. placebo, 5.4.7 Wakefulness-Promoting Medications for Patients with 1-month washout, 3-months melatonin vs. placebo (cross- ISWRD over), and 1-month medication free. Melatonin dosing (using There was no recommendation in the previous Practice Pa- a “standard release” formulation administered 30–40 minutes rameters regarding wakefulness-promoting medications as a before planned bedtime) started at 2 mg, and parents were treatment for patients with ISWRD, and no new studies were given the option of increasing the dose by 2 mg every 3 days, published. until 50% or more improvement in sleep was observed, up There is no evidence to support the use of wakefulness- to 10 mg. The average final dose during the melatonin arm promoting medications as a treatment for patients with IS- was 7 ± 3.0 mg (range 2–10 mg). Parents’ subjective reports of WRD. No recommendation. ISL and TST were improved, although both values crossed the threshold of the predetermined clinically significant minimal 5.4.8 Other Somatic Interventions for Patients with ISWRD change (mean difference −51.71 minutes [CI −13.49 to −89.93] There was no recommendation in the previous Practice Pa- and 48.45 minutes [CI 6.29–90.61 minutes], respectively) (see rameters regarding other somatic interventions as a treatment Table 2). for patients with ISWRD, and no subsequent relevant studies Studies that were not eligible for analysis may nonetheless have been published. provide additional relevant clinical information. In an open- There is no evidence to support the use of other somatic label observational trial in five children with severe develop- interventions for the treatment of patients with ISWRD. mental disabilities and disrupted sleep-wake patterns, Pillar No recommendation. and colleagues153 found that 3 mg melatonin administered at 18:30 increased actigraphically recorded TST. Other trials 5.4.9 Combination Treatments in Demented Elderly Adults have tested the effects of melatonin for nocturnal sleep distur- with ISWRD bance in children with a range of neurodevelopmental difficul- This recommendation was designated as a GUIDELINE ties.154–156 Many of the children included in these studies had in the previous Practice Parameters,1 based upon two stud- at least one comorbid disability (e.g., epilepsy, blindness), and ies, neither of which were included in the present analysis, ei- all had disturbed sleep-wake patterns. Although considerable ther because participants did not clearly have a diagnosis of inter-individual variability was observed in the response to ISWRD158 or because predefined CRITICAL outcomes were melatonin, all of the studies showed significant improvements not measured.159 One randomized controlled trial pertaining to in at least one sleep measure. Another important finding from the treatment of demented elderly patients with ISWRD146 was these studies is the relative safety of melatonin in this popula- published subsequent to 2007, and did not demonstrate benefit tion as no significant side effects were observed. with respect to predefined CRITICAL outcomes. The level of reviewed evidence was VERY LOW (Appendix, Table S12). 5.4.6.2a The TF suggests that clinicians use strategically Dowling and colleagues146 examined sleep-related out- timed melatonin as a treatment for ISWRD in children/ comes from the combination of light treatment (> 2500 lux adolescents with neurologic disorders (versus no white broad spectrum light delivered by light boxes at a dis- treatment). [WEAK FOR] tance of 30–34 inches from the eye between 09:30–10:30 for 10 Summary: This recommendation was designated as an OP- weeks) and either melatonin (5 mg immediate-release between TION in the 2007 Practice Parameters, but none of the reviewed 17:00–18:00) or placebo among 32 nursing home patients with studies were eligible for the current analysis. One subsequently Alzheimer’s disease (mean age 86 ± 8 years). The dim-light published eligible study was identified, with a MODERATE control condition was exposed to indoor light of 150–200 lux level of evidence157 (Appendix, Table S11). The data indicate (see additional study details in section 5.4.4.1 in this paper). that melatonin administration of 2–10 mg during the hour be- The intervention did not significantly improve actigraphically fore planned bedtime may improve CRITICAL sleep outcomes estimated TST, the sole investigated CRITICAL outcome. in children/adolescents with neurologic disorders and ISWRD; Various other studies were ineligible for the current analy- although confidence intervals associated with positive values sis, but nonetheless bear potential relevance to clinicians. Two crossed the threshold of the predetermined clinically signifi- non-blinded, randomized trials examined multimodal treat- cant minimal change (see Table 2). Another caveat is that this ments that included daytime activity, bright light exposure, recommendation is culled from a small sample of patients and sleep scheduling in community-dwelling elderly patients with a range of developmental disorders. As such, it may not with dementia.158,160 The results demonstrated significant de- generalize to all children/adolescents with ISWRD/neurologic creases in nighttime wakefulness, and greater adherence to the disorders. Although no serious adverse reactions have been intervention was associated with more improvement. Combi- described in relation to melatonin use to date, relevant con- nation treatments involving prescribed sleep-wake scheduling, cerns have been raised by select studies with respect to the light exposure and increased daytime activity have also been pediatric/adolescent population,94 and rigorous long-term data examined in two 5-day studies among demented, elderly nurs- are lacking (see separate “Harms and Adverse Effects” sec- ing home residents.159,161 Participants were required to be out of tion). Nevertheless, clinical experience suggests that a majority bed from 08:00 to 20:00, had scheduled low-intensity physical of patients and caregivers would accept this treatment option activity 3 times per day, and exposure to at least 30 minutes of (versus no treatment), particularly taking into account signifi- outdoor sunlight daily. Other procedures included implemen- cant burdens associated with the neurologic disabilities and tation of a structured nighttime routine by caregivers, and min- severe associated sleep disturbances. imization of nighttime noise and interventions. Although these

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1220 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders studies did not measure outcomes defined by the TF, the multi- is designed to be more clinically relevant, as the GRADE sys- modal interventions significantly improved daytime function- tem requires a combined consideration of strength of evidence, ing and amplitude of rest-activity rhythms. The investigations in conjunction with risk/benefit analyses and determination were restricted by constraints inherent to the nursing home of patient values and preferences. As such, many previously environment (e.g., high dropout rate, inability to blind raters endorsed practice recommendations have been negated, and to condition), yet both had relatively large sample sizes. An numerous PICO questions remain unanswered. While this cer- additional caveat is that many more participants were screened tainly points out the significant gaps with respect to the clinical than were eligible for participation, such that the results may CRSWD research (highlighted by the prior Standards of Prac- not generalize to all patients with dementia. tice group),1 these updated Practice Parameters are intended to provide clinicians with heightened confidence in prescrib- 5.4.9.1a The TF suggests that clinicians avoid the use ing treatments and, equally importantly, they should serve as of combined treatments consisting of light therapy in a roadmap for future studies that will propel higher quality, combination with melatonin in demented, elderly patients more sophisticated CRSWD therapies. with ISWRD (versus no treatment). [WEAK AGAINST] Generally speaking, larger more rigorously designed stud- Summary: This recommendation was designated as a ies (randomized placebo-controlled trials) with ICSD-3 de- GUIDELINE in the previous Practice Parameters. One rel- fined CRSWDs5 are required, and replication of results from evant randomized controlled trial146 was published subsequent separate centers is essential. Clinical research pertaining to to 2007. The level of reviewed evidence from this single study ASWPD and ISWRD in particular has suffered significantly was VERY LOW (Appendix, Table S12). Including melatonin due to a lack of adherence to International Classification of as part of a combination treatment with light therapy does not Sleep Disorders diagnostic criteria. In the specific case of AS- appear to confer additional benefit146 and may increase the po- WPD, one can only conjecture that the results of light therapy tential for harms.147 Clinical experience suggests that patients/ studies that address sleep maintenance/early-morning awak- caregivers would carefully consider the risks of depression ening difficulties are applicable to the treatment of ASWPD, and withdrawn behaviors with treatments that include melato- and uncertainty will remain unless strict terminology are used. nin. Thus, the majority of patients/caregivers would not accept Others have raised concerns about the frequency at which one combination treatments consisting of melatonin and bright actually encounters patients with this condition in the clinical light (versus no treatment). Other combination treatments (e.g., arena,163 a topic that is beyond the scope of this discussion. As bright light, scheduled sleep-wake, and physical activity) are for ISWRD, the term is rarely used in the medical literature, worthy of further investigation. presumably because the comorbid disorders that frequently overlap (e.g., dementia, developmental disabilities) tend to 5.4.9.2 Combination treatments for children/adolescents overshadow the CRSWD. Nevertheless, sleep disturbances are with ISWRD among the most onerous of difficulties for caregivers of these The previous Practice Parameters recommendation was des- patients. Thus, lack of consideration of the formal diagnosis of ignated at the OPTION level, based upon the results of one ISWRD in studies of sleep in these populations makes it dif- study.162 No new studies were identified. The previously cited ficult to identify effective treatments for this important clinical investigation was an open trial in children with moderate-to- problem. severe mental retardation and associated nocturnal sleep dis- More specifically, future studies could advance the field by turbances, and employed combination treatment with light including detailed therapeutic information, such as the method therapy, prescribed sleep-wake schedules, and timed daytime and means of treatment delivery (e.g., protective eyewear vs. vo- activity.162 Five of 14 patients showed improvements in TST litional avoidance of light, light therapy intensity/wavelength/ measured on sleep diaries (completed by parents). However, as proximity, continuous versus pulsed light administration [in- it was not clear that these patients met diagnostic criteria for cluding gradually versus abruptly changing illumination164], ISWRD, this study was not eligible for analysis in the present melatonin formulation, etc.), relationship of treatment timing review. with respect to a defined physiologic circadian phase marker or There is insufficient evidence to support the use of com- other sleep parameter, inclusion/exclusion of prescribed sleep/ bination treatments in children/adolescents with ISWRD wake schedules or other behavioral interventions, and study en- (versus no treatment). No recommendation. vironment (laboratory vs. non-laboratory). Regarding the latter factor, field-based studies are sorely needed, and one must be 6.0 CONCLUSIONS AND FUTURE DIRECTIONS cautious not to let tightly controlled bench research prema- turely dictate clinical treatment. As a prime example, there are Circadian-based basic science developments continue to currently no data to support devices that solely deliver blue outpace clinical research pertaining to CRSWDs. Since pub- short wavelength light in the treatment of CRSWDs, and two lication of the prior Practice Parameters,1 relatively few new laboratory-based studies that describe no additional benefit studies have emerged, although it is encouraging that inves- with blue-enriched bright light,24,25 despite the fact that these tigations specifically oriented to the pediatric/adolescent and wavelengths have been identified as especially important for other “special” populations have been published. The major circadian phase resetting in non-clinical experiments.2 More change with the present review is the use of the GRADE sys- importantly, compliance with post-awakening “light boxes” in tem of analysis. While more rigorous in many respects than the the field is very poor,98 and studies that examine the bypassing previously employed evidence-based assessment, derived data of this compliance barrier are particularly intriguing.28,99,165,166

1221 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

Future research should address “dose” of light including lux Related to long-term risks of circadian-based interventions level and duration,17 and should also consider season167 and in general, research is needed to determine the minimum re- other environmental factors that affect overall light exposure quired duration of specific treatments (or to determine that history.19 Finally, more such studies specifically targeting they are required indefinitely), and/or to determine mainte- CRSWD populations are desired. nance treatment schedules. Further studies that investigate From the standpoint of outcomes, similar clinically relevant multi-modal or combination therapies are needed to determine sleep-related measures will be required for inter-study com- whether combinations may prove to be synergistic, and/or to parative purposes (PSG vs. actigraphy vs. subjective reports, extricate independent effects of treatment modalities, so that physiologic or nonphysiologic circadian marker), along with relative successes and failures can be exploited for differ- systematic measures of treatment compliance, to accurately ing clinical scenarios. With respect to the latter point, in the inform clinical practice. In the instance of ISWRD, it should previously cited Gradisar study (involving adolescents with be determined whether separate outcome measures (e.g., circa- DSWPD),104 light therapy was discontinued (and apparently dian amplitude, rest-activity cycle variations) may be superior not required) once a target wake time was reached, at which indicators of treatment efficacy. For all CRSWDs, measures time solely behavioral interventions ensued. It is not clear to of daytime functioning/alertness could enhance clinical what degree this treatment could be generalized to all DSWPD relevance. populations. Inter-study medication comparisons will require equivalent Demonstration of superiority (or lack thereof) of circadian dosing (analyzed melatonin study doses ranged from 0.3–10.0 versus clock-hour TOA for interventions should engender mg), timing (with respect to clock time, typical sleep onset studies that aim to explore demonstrable benefits of phase as- time or other physiologic/non-physiologic circadian marker), sessments in the clinical setting, which in turn could serve to and treatment durations, to accurately gauge benefit. This delineate relative chronobiotic versus hypnotic effects of medi- guideline does not address specific medication/supplement cations/supplements. Some of the reviewed interventions dem- doses (but rather states those cited in the literature). Optimal onstrated successful sleep-related outcomes without changes dosing may vary between CRSWDs. Furthermore, significant in the circadian phase marker and vice versa. In the instance inter-individual differences may also exist within singular that the importance of circadian TOA is demonstrated, it will CRSWDs, as exemplified by a study of N24SWD subjects that be necessary to determine light and melatonin phase PRCs for showed a tenfold variation in entraining doses.136 The issue of adult populations afflicted with CRSWDs (as they may dif- formulation may also be relevant in melatonin studies (regu- fer from normal populations15,172) and to determine the same lar- vs. sustained-release vs. sublingual, etc.), and one group for both afflicted and healthy pediatric/adolescent popula- suggested that slow exogenous melatonin metabolism could be tions. Complicating matters, alterations in phase relationships responsible for a lack of sustained effect in select instances.168 between the circadian timing system and the timing of sleep Taking into account melatonin safety concerns (particularly among those with CRSWDs could impact the ability of inter- among children and those of reproductive age), future properly ventions to exert benefits, even with knowledge of the PRC. powered studies should be performed to identity the lowest For example, longer intervals from various endogenous mela- 173 174–176 effective melatonin dosage and duration of treatment (acute tonin parameters and CBTmin to sleep offset have fre- and maintenance). Long-term physiologic studies are needed quently been described among adult patients with DSWPD to accurately ascertain any serious chronic risks, particularly as compared to controls. However, this finding has not been as melatonin supplements are not subject to FDA oversight.94 demonstrated among protocols in which subjects are forced to In January of 2014, the FDA approved the melatonin ago- maintain a more conventional sleep/wake schedule,27,177,178 sug- nist HetliozTM (tasimelteon) for the treatment of N24SWD gesting that this observation may simply be a consequence of among the blind. This is the first FDA-approved drug for any longer habitual TST. Greater elucidation is required. On a sep- CRSWD. Since the last date of our latest literature search, the arate note, effective treatments may need to address concomi- results of a peer-reviewed, randomized, placebo-controlled tant impairment of homeostatic sleep processes in CRSWDs, study were published.169 This is the largest treatment study as has been demonstrated in DSWPD and among adolescents with N24SWD patients, and it demonstrated that Tasimelteon in general.179,180 Whether hypnotics have a role in this setting safely and effectively treated afflicted patients: entrainment deserves to be further explored.89 occurred in 20% (8 of 40) of patients receiving the drug com- Present guidelines predominantly reflect biological under- pared with 3% (1 of 38) receiving placebo. This entrainment pinnings associated with CRSWDs. Studies are needed to rate is lower than the 67% entrainment rate (12 of 18) found in investigate and understand predominant exogenous and en- the current meta-analysis of melatonin treatment in N24SWD, dogenous contributors to the development and perpetuation but this may have been a product of the short duration of treat- of CRSWDs, so that different subtypes (and possibly differ- ment with Tasimelteon prior to the assessment of entrain- ent treatment/prophylactic regimens) can be identified. In the ment, as noted by the authors. Consistent with this hypothesis, case of adolescents/young adults (and, to a lesser degree, other higher rates of entrainment were found during longer, open- adults181), numerous exogenous factors, such as increased auton- label treatment with tasimelteon. Direct comparisons between omy with respect to sleep time, employment, and involvement Tasimelteon and melatonin have yet to be conducted. At least in extracurricular activities have been identified as variables two other investigations (involving ramelteon) also sug- contributing to the generally observed delay in sleep/wake pat- gest a potential future CRSWD treatment role for melatonin terns,182 but have not been studied among adolescent DSWPD agonists.170,171 cohorts specifically.183,184 Additionally, repeated exposure to

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1222 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders frustrations at not being able to fall asleep at a desired time ABBREVIATIONS can lead to the development of a concomitant conditioned insomnia, which can perpetuate sleep difficulties. Exposure to ADHD, attention deficit hyperactivity disorder indoor lighting during evening hours185–188 and/or delays in aMT6s, 6-sulfatoxymelatonin (urinary metabolite of weekend wake times189–191 have also been implicated as con- melatonin) tributors to persistently delayed sleep/wake times, but have not CBTMin, core body temperature minimum been specifically implicated in adolescent DSWPD.192 Some DLMO, dim light melatonin onset have urged that school lighting environments be optimized for GRADE, grading of recommendations assessment, maximal circadian benefits.193 In the case of N24SWD, it may development and evaluation be that the exogenous and endogenous contributors to the dis- ISL, initial sleep latency order differ between blind and sighted individuals and that this PSG, polysomnography may similarly necessitate different treatment regimens. Iden- SOT, sleep onset time tification and manipulation of exogenous variables in trials of SOffT, sleep offset time CRSWDs may prove fruitful. TF, task force The associated development of clinical profiles would en- TOA, time of administration able clinicians to better ascertain which patients might respond TST, total sleep time to suggested treatments, and related research is encouraged. 104 In the Gradisar study involving adolescents with DSWPD, REFERENCES school non-attendance, unrestricted sleep during vacation periods, and (not surprisingly) amotivation were all noted to be 1. Morgenthaler TI, Lee-Chiong T, Alessi C, et al. Practice parameters for the clinical evaluation and treatment of circadian rhythm sleep disorders. An barriers to successful outcomes with light therapy. Patients fit- American Academy of Sleep Medicine report. Sleep 2007;30:1445–59. ting this profile are perhaps better suited to less complex inter- 2. Sack RL, Auckley D, Auger RR, et al. Circadian rhythm sleep disorders: part ventions. In a separate study involving young adult subjects I, basic principles, shift work and jet lag disorders. An American Academy of with DSWPD and N24SWD receiving melatonin, a higher Sleep Medicine review. Sleep 2007;30:1460–83. 3. Sack RL, Auckley D, Auger RR, et al. Circadian rhythm sleep disorders: part response rate correlated indirectly with shorter habitual TST, II, advanced sleep phase disorder, delayed sleep phase disorder, free-running 131 as well as a later age of onset. Information such as this may disorder, and irregular sleep-wake rhythm. An American Academy of Sleep eventually allow clinicians to optimally tailor treatment. Medicine review. Sleep 2007;30:1484–501. In select cases, accommodation to a CRSWD patient’s cir- 4. American Academy of Sleep Medicine. International classification of sleep disorders, 2nd ed. Westchester, IL: American Academy of Sleep Medicine, cadian preference may be most practical, and further studies 2005. examining implementation of such schedules are desirable. 5. American Academy of Sleep Medicine. International classification of sleep Believing that some CRSWD cases are refractory to treat- disorders, 3rd ed. Darien, IL: American Academy of Sleep Medicine, 2014. ment, Dagan and Abadi (2001)194 recommended foregoing 6. Borbely AA. A two process model of sleep regulation. Hum Neurobiol 1982;1:195–204. therapy (specifically among DSWPD patients), and instead 7. Moore RY, Eichler VB. Loss of a circadian adrenal corticosterone rhythm urged implementation of rehabilitation and accommodation to following suprachiasmatic lesions in the rat. Brain Res 1972;42:201–6. the preferred sleep/wake schedule in select instances, includ- 8. Stephan FK, Zucker I. Circadian rhythms in drinking behavior and locomotor ing support for disability from duties that require strict sleep/ activity of rats are eliminated by hypothalamic lesions. Proc Natl Acad Sci U S A 1972;69:1583–6. wake schedules, and encouragement to pursue endeavors with 9. Ralph MR, Foster RG, Davis FC, Menaker M. Transplanted suprachiasmatic more flexible scheduling. The benefits of such accommodation nucleus determines circadian period. Science 1990;247:975–8. were demonstrated in a separate military-based study, with 10. Welsh DK, Logothetis DE, Meister M, Reppert SM. Individual neurons evidence of superior performance and mood among those en- dissociated from rat suprachiasmatic nucleus express independently phased circadian firing rhythms. Neuron 1995;14:697–706. abled to adapt a relatively delayed sleep/wake schedule, which 11. Czeisler CA, Duffy JF, Shanahan TL, et al. Stability, precision, and near-24- 195 correlated with increased TST. A later school start time may hour period of the human circadian pacemaker. Science 1999;284:2177–81. be sought for adolescents, if practical and available. This inter- 12. Burgess HJ, Eastman CI. Human tau in an ultradian light-dark cycle. J Biol vention alone can significantly increase TST and mitigate as- Rhythms 2008;23:374–6. 196–199 13. Dunlap JC, Loros JL, DeCoursey PJ. Chronobiology: biological timekeeping. sociated impairments. Unfortunately, the implementation Sinauer Associates, Inc., 2004. of this policy change frequently encounters staunch political 14. Lewy AJ, Wehr TA, Goodwin FK, Newsome DA, Markey SP. Light suppresses resistance and is presently available in select regions only.200 melatonin secretion in humans. Science 1980;210:1267–9. In sum, although much work remains, significant progress 15. St Hilaire MA, Gooley JJ, Khalsa SB, Kronauer RE, Czeisler CA, Lockley SW. Human phase response curve to a 1 h pulse of bright white light. J Physiol has been made in the recognition/treatment of CRSWDs since 2012;590:3035–45. the inception of Sleep Medicine as a distinct medical disci- 16. Zeitzer JM, Dijk DJ, Kronauer R, Brown E, Czeisler C. Sensitivity of the pline. Our aim with the present guidelines is to provide clini- human circadian pacemaker to nocturnal light: melatonin phase resetting and cians with immediate access to up-to-date information in order suppression. J Physiol 2000;526 Pt 3:695–702. 17. Chang AM, Santhi N, St Hilaire M, et al. Human responses to bright light of to make properly informed treatment decisions. The circadian different durations. J Physiol 2012;590:3103–12. system impacts a wide range of physiology, and treatment of 18. Hebert M, Martin SK, Lee C, Eastman CI. The effects of prior light history on CRSWDs would therefore be expected to exert a wide-ranging the suppression of melatonin by light in humans. J Pineal Res 2002;33:198– impact on manifold aspects of health. This publication should 203. 19. Chang AM, Scheer FA, Czeisler CA. The human circadian system adapts to serve as an impetus to address clinical research deficiencies prior photic history. J Physiol 2011;589:1095–102. and to promote novel inquiries for treatments of these challenging and interesting conditions.

1223 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

20. Buxton OM, L’Hermite-Baleriaux M, Turek FW, van Cauter E. Daytime naps in 49. Werneke U, Turner T, Priebe S. Complementary medicines in psychiatry: darkness phase shift the human circadian rhythms of melatonin and thyrotropin review of effectiveness and safety. Br J Psychiatry 2006;188:109–21. secretion. Am J Physiol Regul Integr Comp Physiol 2000;278:R373–82. 50. Rubio-Sastre P, Scheer FA, Gomez-Abellan P, Madrid JA, Garaulet M. Acute 21. Burgess HJ, Molina TA. Home lighting before usual bedtime impacts circadian melatonin administration in humans impairs glucose tolerance in both the timing: a field study. Photochem Photobiol 2014;90:723–6. morning and evening. Sleep 2014;37:1715–9. 22. Thapan K, Arendt J, Skene DJ. An action spectrum for melatonin suppression: 51. Seabra ML, Bignotto M, Pinto LR Jr., Tufik S. Randomized, double-blind evidence for a novel non-rod, non-cone photoreceptor system in humans. clinical trial, controlled with placebo, of the toxicology of chronic melatonin J Physiol 2001;535:261–7. treatment. J Pineal Res 2000;29:193–200. 23. Brainard GC, Hanifin JP, Greeson JM, et al. Action spectrum for melatonin 52. Hoebert M, van der Heijden KB, van Geijlswijk IM, Smits MG. Long-term regulation in humans: evidence for a novel circadian photoreceptor. J Neurosci follow-up of melatonin treatment in children with ADHD and chronic sleep 2001;21:6405–12. onset insomnia. J Pineal Res 2009;47:1–7. 24. Smith MR, Eastman CI. Phase delaying the human circadian clock with blue- 53. Wasdell MB, Jan JE, Bomben MM, et al. A randomized, placebo-controlled trial enriched polychromatic light. Chronobiol Int 2009;26:709–25. of controlled release melatonin treatment of delayed sleep phase syndrome 25. Smith MR, Revell VL, Eastman CI. Phase advancing the human circadian and impaired sleep maintenance in children with neurodevelopmental clock with blue-enriched polychromatic light. Sleep Med 2009;10:287–94. disabilities. J Pineal Res 2008;44:57–64. 26. RevMan 5.2. The Nordic Cochrane Center, The Cochrane Collaboration. 54. Carr R, Wasdell MB, Hamilton D, et al. Long-term effectiveness outcome of Copenhagen, 2012. melatonin therapy in children with treatment-resistant circadian rhythm sleep 27. Mundey K, Benloucif S, Harsanyi K, Dubocovich ML, Zee PC. Phase- disorders. J Pineal Res 2007;43:351–9. dependent treatment of delayed sleep phase syndrome with melatonin. Sleep 55. Valcavi R, Zini M, Maestroni GJ, Conti A, Portioli I. Melatonin stimulates 2005;28:1271–8. growth hormone secretion through pathways other than the growth hormone- 28. Cole RJ, Smith JS, Alcala YC, Elliott JA, Kripke DF. Bright-light mask treatment releasing hormone. Clin Endocrinol 1993;39:193–9. of delayed sleep phase syndrome. J Biol Rhythms 2002;17:89–101. 56. Luboshitzky R, Shen-Orr Z, Nave R, Lavi S, Lavie P. Melatonin administration 29. van Geijlswijk IM, van der Heijden KB, Egberts AC, Korzilius HP, Smits alters semen quality in healthy men. J Androl 2002;23:572–8. MG. Dose finding of melatonin for chronic idiopathic childhood sleep onset 57. van Geijlswijk IM, Mol RH, Egberts TC, Smits MG. Evaluation of sleep, insomnia: an RCT. Psychopharmacology 2010;212:379–91. puberty and mental health in children with long-term melatonin treatment 30. Guyatt G, Oxman AD, Akl EA, et al. GRADE guidelines: 1. Introduction- for chronic idiopathic childhood sleep onset insomnia. Psychopharmacology GRADE evidence profiles and summary of findings tables. J Clin Epidemiol 2011;216:111–20. 2011;64:383–94. 58. Ilomaki J, Paljarvi T, Korhonen MJ, et al. Prevalence of concomitant use of 31. Balshem H, Helfand M, Schunemann HJ, et al. GRADE guidelines: 3. Rating alcohol and sedative-hypnotic drugs in middle and older aged persons: a the quality of evidence. J Clin Epidemiol 2011;64:401–6. systematic review. Ann Pharmacother 2013;47:257–68. 32. Guyatt GH, Oxman AD, Vist G, et al. GRADE guidelines: 4. Rating the quality 59. Buscemi N, Vandermeer B, Friesen C, et al. The efficacy and safety of drug of evidence--study limitations (risk of bias). J Clin Epidemiol 2011;64:407–15. treatments for chronic insomnia in adults: a meta-analysis of RCTs. J Gen 33. Guyatt GH, Oxman AD, Montori V, et al. GRADE guidelines: 5. Rating the Intern Med 2007;22:1335–50. quality of evidence--publication bias. J Clin Epidemiol 2011;64:1277–82. 60. Peron EP, Gray SL, Hanlon JT. Medication use and functional status decline 34. Guyatt GH, Oxman AD, Kunz R, et al. GRADE guidelines 6. Rating the quality in older adults: a narrative review. Am J Geriatr Pharmacother 2011;9:378–91. of evidence--imprecision. J Clin Epidemiol 2011;64:1283–93. 61. Woolcott JC, Richardson KJ, Wiens MO et al. Meta-analysis of the impact 35. Guyatt GH, Oxman AD, Kunz R, et al. GRADE guidelines: 7. Rating the quality of 9 medication classes on falls in elderly persons. Arch Intern Med of evidence--inconsistency. J Clin Epidemiol 2011;64:1294–302. 2009;169:1952–60. 36. Guyatt GH, Oxman AD, Kunz R, et al. GRADE guidelines: 8. Rating the quality 62. Bakken MS, Engeland A, Engesaeter LB, Ranhoff AH, Hunskaar S, Ruths S. of evidence--indirectness. J Clin Epidemiol 2011;64:1303–10. Risk of hip fracture among older people using anxiolytic and hypnotic drugs: 37. Guyatt GH, Oxman AD, Sultan S, et al. GRADE guidelines: 9. Rating up the a nationwide prospective cohort study. Eur J Clin Pharmacol 2014;70:873–80. quality of evidence. J Clin Epidemiol 2011;64:1311–6. 63. Deschenes CL, McCurry SM. Current treatments for sleep disturbances in 38. Tuunainen A, Kripke DF, Endo T. Light therapy for non-seasonal depression. individuals with dementia. Curr Psychiatry Reps 2009;11:20–6. Cochrane Database System Rev 2004:CD004050. 64. Hajak G, Muller WE, Wittchen HU, Pittrow D, Kirch W. Abuse and dependence 39. Dauphinais DR, Rosenthal JZ, Terman M, DiFebo HM, Tuggle C, Rosenthal potential for the non-benzodiazepine hypnotics zolpidem and zopiclone: a NE. Controlled trial of safety and efficacy of bright light therapy vs. negative review of case reports and epidemiological data. Addiction 2003;98:1371–8. air ions in patients with bipolar depression. Psychiatry Res 2012;196:57–61. 65. James SP, Mendelson WB. The use of trazodone as a hypnotic: a critical 40. Terman M, Terman JS. Light therapy for seasonal and nonseasonal depression: review. J Clin Psychiatry 2004;65:752–5. efficacy, protocol, safety, and side effects. CNS Spectrums 2005;10:647–63; 66. Stone M. Mortality and antipsychotic drug use in dementia-related behavioral quiz 72. disorders. US Department of Health and Human Services, Food and Drug 41. Ulrich V, Olesen J, Gervil M, Russell MB. Possible risk factors and precipitants Administration, Center for Drug Evaluation and Research, 2005. for migraine with aura in discordant twin-pairs: a population-based study. 67. Moldofsky H, Musisi S, Phillipson EA. Treatment of a case of advanced sleep Cephalalgia 2000;20:821–5. phase syndrome by phase advance chronotherapy. Sleep 1986;9:61–5. 42. Gallin PF, Terman M, Reme CE, Rafferty B, Terman JS, Burde RM. 68. Suhner AG, Murphy PJ, Campbell SS. Failure of timed bright light exposure Ophthalmologic examination of patients with seasonal affective disorder, to alleviate age-related sleep maintenance insomnia. J Am Geriatr Soc before and after bright light therapy. Am J Ophthalmol 1995;119:202–10. 2002;50:617–23. 43. Reme CE, Rol P, Grothmann K, Kaase H, Terman M. Bright light therapy 69. Pallesen S, Nordhus IH, Skelton SH, Bjorvatn B, Skjerve A. Bright light in focus: lamp emission spectra and ocular safety. Technol Health Care treatment has limited effect in subjects over 55 years with mild early morning 1996;4:403–13. awakening. Percept Motor Skills 2005;101:759–70. 44. Ferracioli-Oda E, Qawasmi A, Bloch MH. Meta-analysis: melatonin for the 70. Lack L, Wright H, Kemp K, Gibbon S. The treatment of early-morning treatment of primary sleep disorders. PloS One 2013;8:e63773. awakening insomnia with 2 evenings of bright light. Sleep 2005;28:616–23. 45. Herxheimer A, Petrie KJ. Melatonin for the prevention and treatment of jet lag. 71. Lack L, Wright H. The effect of evening bright light in delaying the circadian Cochrane Database Systematic Rev 2002:CD001520. rhythms and lengthening the sleep of early morning awakening insomniacs. 46. Spadoni G, Bedini A, Rivara S, Mor M. Melatonin receptor agonists: new options Sleep 1993;16:436–43. for insomnia and depression treatment. CNS Neurosci Ther 2011;17:733–41. 72. Palmer CR, Kripke DF, Savage HC Jr., Cindrich LA, Loving RT, Elliott JA. 47. Committee on the Framework for Evaluating the Safety of the Dietary Efficacy of enhanced evening light for advanced sleep phase syndrome. Supplements FaNB, Board on Life Sciences, Institute of Medicine and Behav Sleep Med 2003;1:213–26. National Research Council of the National Academies. Dietary supplements: 73. Campbell SS, Dawson D, Anderson MW. Alleviation of sleep maintenance a framework for evaluating safety. Washington, DC: The National Academies insomnia with timed exposure to bright light. J Am Geriatr Society Press, 2005. 1993;41:829–36. 48. Buscemi N, Vandermeer B, Hooton N, et al. The efficacy and safety of 74. Munch M, Scheuermaier KD, Zhang R, et al. Effects on subjective and exogenous melatonin for primary sleep disorders. A meta-analysis. J Gen objective alertness and sleep in response to evening light exposure in older Intern Med 2005;20:1151–8. subjects. Behav Brain Res 2011;224:272–8.

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1224 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

75. Ito A, Ando K, Hayakawa T, et al. Long-term course of adult patients with 100. Revell VL, Burgess HJ, Gazda CJ, Smith MR, Fogg LF, Eastman CI. Advancing delayed sleep phase syndrome. Jpn J Psychiatry Neurol 1993;47:563–7. human circadian rhythms with afternoon melatonin and morning intermittent 76. Czeisler CA, Richardson GS, Coleman RM, et al. Chronotherapy: resetting bright light. J Clin Endocrinol Metab 2006;91:54–9. the circadian clocks of patients with delayed sleep phase insomnia. Sleep 101. Burke TM, Markwald RR, Chinoy ED, et al. Combination of light and 1981;4:1–21. melatonin time cues for phase advancing the human circadian clock. Sleep 77. Oren DA, Wehr TA. Hypernyctohemeral syndrome after chronotherapy for 2013;36:1617–24. delayed sleep phase syndrome. N Engl J Med 1992;327:1762. 102. Paul MA, Gray GW, Lieberman HR. et al. Phase advance with separate 78. Sharkey KM, Carskadon MA, Figueiro MG, Zhu Y, Rea MS. Effects of an and combined melatonin and light treatment. Psychopharmacology advanced sleep schedule and morning short wavelength light exposure 2011;214:515–23. on circadian phase in young adults with late sleep schedules. Sleep Med 103. Samaranayake CB, Fernando A, Warman G. Outcome of combined melatonin 2011;12:685–92. and bright light treatments for delayed sleep phase disorder. Royal Aust NZ J 79. Iwamitsu Y, Ozeki Y, Konishi M, Murakami J, Kimura S, Okawa M. Psychological Psychiatry 2010;44:676. characteristics and the efiicacy of hospitalization treatment on delayed sleep 104. Gradisar M, Dohnt H, Gardner G, et al. A randomized controlled trial of phase syndrome patients with school refusal. Sleep Biol Rhythms 2007;5:15– cognitive-behavior therapy plus bright light therapy for adolescent delayed 22. sleep phase disorder. Sleep 2011;34:1671–80. 80. de Sousa IC, Araújo IC, de Azevedo CV. The effect of a sleep hygiene 105. Sack RL, Lewy AJ, Blood ML, Keith LD, Nakagawa H. Circadian rhythm education program on the sleep-wake cycle of Brazilian adolescent students. abnormalities in totally blind people: incidence and clinical significance. J Clin Sleep Biol Rhythms 2007;5:251–58. Endocrinol Metab 1992;75:127–34. 81. Fargason RE, Preston T, Hammond E, May R, Gamble KL. Treatment of 106. Czeisler CA, Shanahan TL, Klerman EB, et al. Suppression of melatonin attention deficit hyperactivity disorder insomnia with blue wavelength light- secretion in some blind patients by exposure to bright light. N Engl J Med blocking glasses. Chronophysiol Ther 2013:1–8. 1995;332:6–11. 82. Burkhart K, Phelps JR. Amber lenses to block blue light and improve sleep: a 107. Danilenko KV, Cajochen C, Wirz-Justice A. Is sleep per se a zeitgeber in randomized trial. Chronobiol Int 2009;26:1602–12. humans? J Biol Rhythms 2003;18:170–8. 83. Rosenthal NE, Joseph-Vanderpool JR, Levendosky AA, et al. Phase-shifting 108. Gordijn MC, Beersma DG, Korte HJ, van den Hoofdakker RH. Effects of light effects of bright morning light as treatment for delayed sleep phase syndrome. exposure and sleep displacement on dim light melatonin onset. J Sleep Res Sleep 1990;13:354–61. 1999;8:163–74. 84. Watanabe T, Kajimura N, Kato M, Sekimoto M, Takahashi K. Effects of 109. Hoban TM, Lewy AJ, Sack RL, Singer CM. The effects of shifting sleep two phototherapy in patients with delayed sleep phase syndrome. Psychiatry Clin hours within a fixed photoperiod. J Neural Transm 1991;85:61–8. Neurosci 1999;53:231–3. 110. Buxton OM, Lee CW, L’Hermite-Baleriaux M, Turek FW, Van Cauter E. 85. Lack L, Bramwell T, Wright H, Kemp K. Morning blue light can advance the Exercise elicits phase shifts and acute alterations of melatonin that vary with melatonin rhythm in mild delayed sleep phase syndrome. Sleep Biol Rhythms circadian phase. Am J Physiol Regul Integr Comp Physiol 2003;284:R714–24. 2007:78–80. 111. Klerman EB, Rimmer DW, Dijk DJ, Kronauer RE, Rizzo JF 3rd, Czeisler CA. 86. Ohta T, Iwata T, Kayukawa Y, Okada T. Daily activity and persistent sleep- Nonphotic entrainment of the human circadian pacemaker. Am J Physiol wake schedule disorders. Progr Neuropsychopharmacol Biol Psychiatry 1998;274:R991–6. 1992;16:529–37. 112. Oren DA, Giesen HA, Wehr TA. Restoration of detectable melatonin 87. Mizuma H, Miyahara Y, Sakamoto T, Kotorii T, Nakazawa Y. Two cases after entrainment to a 24-hour schedule in a ‘free-running’ man. of delayed sleep phase syndrome (DSPS). Jpn J Psychiatry Neurol Psychoneuroendocrinology 1997;22:39–52. 1991;45:163–64. 113. Hoban TM, Sack RL, Lewy AJ, Miller LS, Singer CM. Entrainment of a free- 88. Yamadera H, Takahashi K, Okawa M. A multicenter study of sleep-wake running human with bright light? Chronobiol Int 1989;6:347–53. rhythm disorders: therapeutic effects of vitamin B12, bright light therapy, 114. Watanabe T, Kajimura N, Kato M, Sekimoto M, Hori T, Takahashi K. Case of a chronotherapy and hypnotics. Psychiatry Clin Neurosci 1996;50:203–9. non-24 h sleep-wake syndrome patient improved by phototherapy. Psychiatry 89. Walsh JK, Deacon S, Dijk DJ, Lundahl J. The selective extrasynaptic Clin Neurosci 2000;54:369–70. GABAA agonist, gaboxadol, improves traditional hypnotic efficacy measures 115. Okawa M, Uchiyama M, Ozaki S, et al. Melatonin treatment for circadian and enhances slow wave activity in a model of transient insomnia. Sleep rhythm sleep disorders. Psychiatry Clin Neurosci 1998;52:259–60. 2007;30:593–602. 116. Hayakawa T, Kamei Y, Urata J, et al. Trials of bright light exposure and 90. Kayumov L, Brown G, Jindal R, Buttoo K, Shapiro CM. A randomized, double- melatonin administration in a patient with non-24 hour sleep-wake syndrome. blind, placebo-controlled crossover study of the effect of exogenous melatonin Psychiatry Clin Neurosci 1998;52:261–2. on delayed sleep phase syndrome. Psychosom Med 2001;63:40–8. 117. Klerman EB, Shanahan TL, Brotman DJ, et al. Photic resetting of the human 91. Dahlitz M, Alvarez B, Vignau J, English J, Arendt J, Parkes JD. Delayed sleep circadian pacemaker in the absence of conscious vision. J Biol Rhythms phase syndrome response to melatonin. Lancet 1991;337:1121–4. 2002;17:548–55. 92. Dagan Y, Yovel I, Hallis D, Eisenstein M, Raichik I. Evaluating the role of 118. Arendt J, Aldhous M, Wright J. Synchronisation of a disturbed sleep-wake melatonin in the long-term treatment of delayed sleep phase syndrome cycle in a blind man by melatonin treatment. Lancet 1988;1:772–3. (DSPS). Chronobiol Int 1998;15:181–90. 119. Folkard S, Arendt J, Aldhous M, Kennett H. Melatonin stabilises sleep onset 93. Rahman SA, Kayumov L, Shapiro CM. Antidepressant action of melatonin in time in a blind man without entrainment of cortisol or temperature rhythms. the treatment of delayed sleep phase syndrome. Sleep Med 2010;11:131–6. Neurosci Lett 1990;113:193–8. 94. Kennaway DJ. Potential safety issues in the use of the hormone melatonin in 120. Lapierre O, Dumont M. Melatonin treatment of a non-24-hour sleep-wake paediatrics. J Paediatr Child Health 2015;51:584–9. cycle in a blind retarded child. Biol Psychiatry 1995;38:119–22. 95. Van der Heijden KB, Smits MG, Van Someren EJ, Ridderinkhof KR, Gunning 121. Tzischinsky O, Pal I, Epstein R, Dagan Y, Lavie P. The importance of timing in WB. Effect of melatonin on sleep, behavior, and cognition in ADHD and chronic melatonin administration in a blind man. J Pineal Res 1992;12:105–8. sleep-onset insomnia. J Am Acad Child Adolesc Psychiatry 2007;46:233–41. 122. Lockley SW, Skene DJ, James K, Thapan K, Wright J, Arendt J. Melatonin 96. Smits MG, Nagtegaal EE, van der Heijden J, Coenen AM, Kerkhof GA. administration can entrain the free-running circadian system of blind subjects. Melatonin for chronic sleep onset insomnia in children: a randomized placebo- J Endocrinol 2000;164:R1–6. controlled trial. J Child Neurol 2001;16:86–92. 123. Sack RL, Brandes RW, Kendall AR, Lewy AJ. Entrainment of free-running 97. Okawa M, Takahashi K, Egashira K, et al. Vitamin B12 treatment for delayed circadian rhythms by melatonin in blind people. N Engl J Med 2000;343:1070–7. sleep phase syndrome: a multi-center double-blind study. Psychiatry Clin 124. Hack LM, Lockley SW, Arendt J, Skene DJ. The effects of low-dose 0.5-mg Neurosci 1997;51:275–9. melatonin on the free-running circadian rhythms of blind subjects. J Biol 98. Bjorvatn B, Pallesen S. A practical approach to circadian rhythm sleep Rhythms 2003;18:420–9. disorders. Sleep Med Rev 2009;13:47–60. 125. Lewy AJ, Bauer VK, Hasler BP, Kendall AR, Pires ML, Sack RL. Capturing the 99. Figueiro MG, Rea MS. Preliminary evidence that light through the eyelids can circadian rhythms of free-running blind people with 0.5 mg melatonin. Brain suppress melatonin and phase shift dim light melatonin onset. BMC Res Notes Res 2001;918:96–100. 2012;5:221. 126. Lewy AJ, Hasler BP, Emens JS, Sack RL. Pretreatment circadian period in free-running blind people may predict the phase angle of entrainment to melatonin. Neurosci Lett 2001;313:158–60.

1225 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

127. Lewy AJ, Emens JS, Sack RL, Hasler BP, Bernert RA. Low, but not high, 152. Singer C, Tractenberg RE, Kaye J, et al. A multicenter, placebo-controlled trial doses of melatonin entrained a free-running blind person with a long circadian of melatonin for sleep disturbance in Alzheimer’s disease. Sleep 2003;26:893– period. Chronobiol Int 2002;19:649–58. 901. 128. Lewy AJ, Emens JS, Lefler BJ, Yuhas K, Jackman AR. Melatonin entrains 153. Pillar G, Shahar E, Peled N, Ravid S, Lavie P, Etzioni A. Melatonin improves free-running blind people according to a physiological dose-response curve. sleep-wake patterns in psychomotor retarded children. Pediatr Neurol Chronobiol Int 2005;22:1093–106. 2000;23:225–8. 129. Lewy AJ, Emens JS, Bernert RA, Lefler BJ. Eventual entrainment of the human 154. Jan JE, Espezel H, Appleton RE. The treatment of sleep disorders with circadian pacemaker by melatonin is independent of the circadian phase of melatonin. Dev Med Child Neurol 1994;36:97–107. treatment initiation: clinical implications. J Biol Rhythms 2004;19:68–75. 155. Jan JE, Hamilton D, Seward N, Fast DK, Freeman RD, Laudon M. Clinical trials 130. Siebler M, Steinmetz H, Freund HJ. Therapeutic entrainment of circadian of controlled-release melatonin in children with sleep-wake cycle disorders. rhythm disorder by melatonin in a non-blind patient. J Neurol 1998;245:327–8. J Pineal Res 2000;29:34–9. 131. Kamei Y, Hayakawa T, Urata J, et al. Melatonin treatment for circadian rhythm 156. McArthur AJ, Budden SS. Sleep dysfunction in Rett syndrome: a trial of sleep disorders. Psychiatry Clin Neurosci 2000;54:381–2. exogenous melatonin treatment. Dev Med Child Neurol 1998;40:186–92. 132. McArthur AJ, Lewy AJ, Sack RL. Non-24-hour sleep-wake syndrome in a 157. Wright B, Sims D, Smart S et al. Melatonin versus placebo in children with sighted man: circadian rhythm studies and efficacy of melatonin treatment. autism spectrum conditions and severe sleep problems not amenable to Sleep 1996;19:544–53. behaviour management strategies: a randomised controlled crossover trial. J 133. Okawa M, Mishima K, Nanami T, et al. Vitamin B12 treatment for sleep-wake Autism Dev Disord 2011;41:175–84. rhythm disorders. Sleep 1990;13:15–23. 158. McCurry SM, Gibbons LE, Logsdon RG, Vitiello MV, Teri L. Nighttime insomnia 134. Ohta T, Ando K, Iwata T, et al. Treatment of persistent sleep-wake schedule treatment and education for Alzheimer’s disease: a randomized, controlled disorders in adolescents with methylcobalamin (vitamin B12). Sleep trial. J Am Geriatr Soc 2005;53:793–802. 1991;14:414–8. 159. Alessi CA, Martin JL, Webber AP, Cynthia Kim E, Harker JO, Josephson KR. 135. Kamgar-Parsi B, Wehr TA, Gillin JC. Successful treatment of human non-24- Randomized, controlled trial of a nonpharmacological intervention to improve hour sleep-wake syndrome. Sleep 1983;6:257–64. abnormal sleep/wake patterns in nursing home residents. J Am Geriatr Soc 136. Sugita Y, Ishikawa H, Mikami A, et al. Successful treatment for a patient with 2005;53:803–10. hypernychthermal syndrome. Sleep Res 1987:642. 160. McCurry SM, Pike KC, Vitiello MV, Logsdon RG, Larson EB, Teri L. Increasing 137. Mishima K, Okawa M, Hishikawa Y, Hozumi S, Hori H, Takahashi K. Morning walking and bright light exposure to improve sleep in community-dwelling bright light therapy for sleep and behavior disorders in elderly patients with persons with Alzheimer’s disease: results of a randomized, controlled trial. dementia. Acta Psychiatr Scand 1994;89:1–7. J Am Geriatr Soc 2011;59:1393–402. 138. Fetveit A, Bjorvatn B. The effects of bright-light therapy on actigraphical 161. Martin JL, Marler MR, Harker JO, Josephson KR, Alessi CA. A multicomponent measured sleep last for several weeks post-treatment. A study in a nursing nonpharmacological intervention improves activity rhythms among nursing home population. J Sleep Res 2004;13:153–8. home residents with disrupted sleep/wake patterns. J Gerontol A Biol Sci Med 139. Ancoli-Israel S, Gehrman P, Martin JL, et al. Increased light exposure Sci 2007;62:67–72. consolidates sleep and strengthens circadian rhythms in severe Alzheimer’s 162. Guilleminault C, McCann CC, Quera-Salva M, Cetel M. Light therapy disease patients. Behav Sleep Med 2003;1:22–36. as treatment of dyschronosis in brain impaired children. Eur J Pediatr 140. Ancoli-Israel S, Martin JL, Kripke DF, Marler M, Klauber MR Effect of light 1993;152:754–9. treatment on sleep and circadian rhythms in demented nursing home patients. 163. Auger RR. Advance-related sleep complaints and advanced sleep phase J Am Geriatr Soc 2002;50:282–9. disorder. Sleep Med Clin 2009;4:219–27. 141. Skjerve A, Holsten F, Aarsland D, Bjorvatn B, Nygaard HA, Johansen IM. 164. Kondo M, Tokura H, Wakamura T, et al. Influences of twilight on diurnal Improvement in behavioral symptoms and advance of activity acrophase after variation of core temperature, its nadir, and urinary 6-hydroxymelatonin sulfate short-term bright light treatment in severe dementia. Psychiatry Clin Neurosci during nocturnal sleep and morning drowsiness. Coll Antropol 2009;33:193–9. 2004;58:343–7. 165. Fromm E, Horlebein C, Meergans A, Niesner M, Randler C. Evaluation of a 142. Dowling GA, Hubbard EM, Mastick J, Luxenberg JS, Burr RL, Van Someren dawn simulator in children and adolescents. Biol Rhythm Res 2011;42:417–25. EJ. Effect of morning bright light treatment for rest-activity disruption in 166. Terman M, Jiuan Su T. Circadian rhythm phase advance with dawn simulation institutionalized patients with severe Alzheimer’s disease. Int Psychogeriatr treatment for winter depression. J Biol Rhythms 2010;25:297–301. 2005;17:221–36. 167. Higuchi S, Motohashi Y, Ishibashi K, Maeda T. Less exposure to daily 143. Van Someren EJ, Kessler A, Mirmiran M, Swaab DF. Indirect bright light ambient light in winter increases sensitivity of melatonin to light suppression. improves circadian rest-activity rhythm disturbances in demented patients. Chronobiol Int 2007;24:31–43. Biol Psychiatry 1997;41:955–63. 168. Braam W, van Geijlswijk I, Keijzer H, Smits MG, Didden R, Curfs LM. Loss of 144. Fetveit A, Skjerve A, Bjorvatn B. Bright light treatment improves sleep in response to melatonin treatment is associated with slow melatonin metabolism. institutionalised elderly--an open trial. Int J Geriatr Psychiatry 2003;18:520–6. J Intellect Disabil Res 2010;54:547–55. 145. Satlin A, Volicer L, Ross V, Herz L, Campbell S. Bright light treatment of 169. Lockley SW, Dressman MA, Licamele L, et al. Tasimelteon for non-24-hour behavioral and sleep disturbances in patients with Alzheimer’s disease. Am J sleep-wake disorder in totally blind people (SET and RESET): two multicentre, Psychiatry 1992;149:1028–32. randomised, double-masked, placebo-controlled phase 3 trials. Lancet 2015 146. Dowling GA, Burr RL, Van Someren EJ, et al. Melatonin and bright-light Aug 4. [Epub ahead of print]. treatment for rest-activity disruption in institutionalized patients with 170. Richardson GS, Zee PC, Wang-Weigand S, Rodriguez L, Peng X. Circadian Alzheimer’s disease. J Am Geriatr Soc 2008;56:239–46. phase-shifting effects of repeated ramelteon administration in healthy adults. 147. Riemersma-van der Lek RF, Swaab DF, Twisk J, Hol EM, Hoogendijk WJ, Van J Clin Sleep Med 2008;4:456–61. Someren EJ. Effect of bright light and melatonin on cognitive and noncognitive 171. Rajaratnam SM, Polymeropoulos MH, Fisher DM, et al. Melatonin agonist function in elderly residents of group care facilities: a randomized controlled tasimelteon (VEC-162) for transient insomnia after sleep-time shift: two trial. JAMA 2008;299:2642–55. randomised controlled multicentre trials. Lancet 2009;373:482–91. 148. Greenblatt DJ, Harmatz JS, Singh NN, et al. Pharmacokinetics of zolpidem 172. Burgess HJ, Revell VL, Molina TA, Eastman CI. Human phase response curves from sublingual zolpidem tartrate tablets in healthy elderly versus non-elderly to three days of daily melatonin: 0.5 mg versus 3.0 mg. J Clin Endocrinol Metab subjects. Drugs Aging 2014;31:731–6. 2010;95:3325–31. 149. Berry SD, Lee Y, Cai S, Dore DD. Nonbenzodiazepine sleep medication use 173. Shibui K, Uchiyama M, Okawa M. Melatonin rhythms in delayed sleep phase and hip fractures in nursing home residents. JAMA Intern Med 2013;173:754– syndrome. J Biol Rhythms 1999;14:72–6. 61. 174. Ozaki S, Uchiyama M, Shirakawa S, Okawa M. Prolonged interval from 150. Hartikainen S, Rahkonen T, Kautiainen H, Sulkava R. The use of psychotropics body temperature nadir to sleep offset in patients with delayed sleep phase and survival in demented elderly individuals. Int Clin Psychopharmacol syndrome. Sleep 1996;19:36–40. 2005;20:227–31. 175. Watanabe T, Kajimura N, Kato M, et al. Sleep and circadian rhythm disturbances 151. Serfaty M, Kennell-Webb S, Warner J, Blizard R, Raven P. Double blind in patients with delayed sleep phase syndrome. Sleep 2003;26:657–61. randomised placebo controlled trial of low dose melatonin for sleep disorders in dementia. Int J Geriatr Psychiatry 2002;17:1120–7.

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1226 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

176. Uchiyama M, Okawa M, Shibui K, et al. Altered phase relation between decisions regarding corresponding amendments in the text were left to the TF’s sleep timing and core body temperature rhythm in delayed sleep phase discretion. The project was funded by the American Academy of Sleep Medicine. syndrome and non-24-hour sleep-wake syndrome in humans. Neurosci Lett 2000;294:101–4. 177. Wyatt JK, Stepanski EJ, Kirkby J. Circadian phase in delayed sleep phase SUBMISSION & CORRESPONDENCE INFORMATION syndrome: predictors and temporal stability across multiple assessments. Sleep 2006;29:1075–80. Submitted for publication August, 2015 178. Chang AM, Reid KJ, Gourineni R, Zee PC. Sleep timing and circadian phase in Accepted for publication August, 2015 delayed sleep phase syndrome. J Biol Rhythms 2009;24:313–21. Address correspondence to: R. Robert Auger, MD, Mayo Center for Sleep Medicine, 179. Jenni OG, Achermann P, Carskadon MA. Homeostatic sleep regulation in Rochester, MN; Email: [email protected] adolescents. Sleep 2005;28:1446–54. 180. Uchiyama M, Okawa M, Shibui K et al. Poor compensatory function for sleep loss as a pathogenic factor in patients with delayed sleep phase syndrome. DISCLOSURE STATEMENT Sleep 2000;23:553–8. This was not an industry supported study. The authors have indicated no financial 181. Asaoka A, Fukuda K, Tsutsui Y, Yamazaki K. Does television viewing cause conflicts of interest. Drs. Deriy and Thomas are employed by the American Acad- delayed and/or irregularsleep–wake patterns? Sleep Biol Rhythms 2007;5:23– emy of Sleep Medicine. 27. 182. Crowley SJ, Acebo C, Carskadon MA. Sleep, circadian rhythms, and delayed phase in adolescence. Sleep Med 2007;8:602–12. 183. Carskadon MA. Patterns of sleep and sleepiness in adolescents. Pediatrician APPENDIX 1990;17:5–12. 184. Wyatt JK. Delayed sleep phase syndrome: pathophysiology and treatment options. Sleep 2004;27:1195–203. 185. Peixoto CA, da Silva AG, Carskadon MA, Louzada FM. Adolescents living in Search Terms and MeSH Used in the Literature homes without electric lighting have earlier sleep times. Behav Sleep Med Search #1 2009;7:73–80. Sleep disorders, circadian rhythm”[MeSH Terms] OR 186. Vollmer C, Michel U, Randler C. Outdoor light at night (LAN) is correlated with (“sleep”[All Fields] AND “disorders”[All Fields] AND eveningness in adolescents. Chronobiol Int 2012;29:502–8. “circadian”[All Fields] AND “rhythm”[All Fields]) 187. Cajochen C, Frey S, Anders D, et al. Evening exposure to a light-emitting diodes (LED)-backlit computer screen affects circadian physiology and OR “circadian rhythm sleep disorders”[All Fields] OR cognitive performance. J Appl Physiol 2011;110:1432–8. (“circadian”[All Fields] AND “rhythm”[All Fields] 188. Figueiro MG, Rea MS. Evening daylight may cause adolescents to sleep less AND “sleep”[All Fields] AND “disorder”[All Fields]) OR in spring than in winter. Chronobiol Int 2010;27:1242–58. “circadian rhythm sleep disorder”[All Fields] OR (free[All 189. Taylor A, Wright HR, Lack LC. Sleeping-in on the weekend delays circadian phase and increases sleepiness the following week. Sleep Biol Rhythms Fields] AND “running”[All Fields] AND “disorder”[All 2008;6:172–79. Fields]) OR (irregular[All Fields] AND sleep-wake[All 190. Burgess HJ, Eastman CI. A late wake time phase delays the human dim light Fields] AND rhythm[All Fields]) OR (non[All Fields] AND melatonin rhythm. Neurosci Lett 2006;395:191–5. 24-hour[All Fields] AND (“sleep disorders”[MeSH Terms] 191. Crowley SJ, Carskadon MA. Modifications to weekend recovery sleep delay circadian phase in older adolescents. Chronobiol Int 2010;27:1469–92. OR (“sleep”[All Fields] AND “disorders”[All Fields]) OR 192. Auger RR, Burgess HJ, Dierkhising RA, Sharma RG, Slocumb NL. Light “sleep disorders”[All Fields] OR (“sleep”[All Fields] AND exposure among adolescents with delayed sleep phase disorder: a prospective “disorder”[All Fields]) OR “sleep disorder”[All Fields])) OR cohort study. Chronobiol Int 2011;28:911–20. ((“Blindness”[Mesh] OR (“blindness”[MeSH Terms] OR 193. Figueiro MG, Rea MS. Lack of short-wavelength light during the school day delays dim light melatonin onset (DLMO) in middle school students. “blindness”[All Fields] OR nonsighted[All Fields])) AND Neuroendocrinol Lett 2010;31:92–6. (“sleep disorders”[MeSH Terms] OR (“sleep”[All Fields] 194. Dagan Y, Abadi J. Sleep-wake schedule disorder disability: a lifelong AND “disorders”[All Fields]) OR “sleep disorders”[All untreatable pathology of the circadian time structure. Chronobiol Int Fields])) AND (Meta-Analysis[ptyp] OR Practice 2001;18:1019–27. 195. Miller NL, Tvaryanas AP, Shattuck LG. Accommodating adolescent sleep-wake Guideline[ptyp] OR systematic[sb]). patterns: the effects of shifting the timing of sleep on training effectiveness. Sleep 2012;35:1123–36. Terms for the Literature Search #2 in Pubmed 196. National Sleep Foundation. 2006 Sleep in America Poll Summary Findings, Washington, DC: National Sleep Foundation, 2006. Database 197. Wahlstrom K. Changing times: findings from the first longitudinal study of later ((((circadian rhythm sleep disorders[tw] OR circadian rhythm school start times. NAASP Bull 2002;86:3–21. sleep disorder[tw] OR “Sleep Disorders, Circadian 198. Danner F, Phillips B. Adolescent sleep, school start times, and teen motor Rhythm”[Mesh:noexp] OR “chronobiology disorders”[tw] OR vehicle crashes. J Clin Sleep Med 2008;4:533–5. “Chronobiology Disorders”[Mesh:noexp] OR 199. Owens JA, Belon K, Moss P. Impact of delaying school start time on adolescent sleep, mood, and behavior. Arch Pediatr Adolesc Med 2010;164:608–14. hypernychthemeral[tw] OR circadian misalignment[tw] OR 200. Wahlstrom K. The prickly politics of school starting times. Phi Delta Kappa circadian dysregulation[tw]) OR ((irregular[tw] OR non 24 1999;80:344–47. hour[tw]) AND (sleep wake[tw] OR sleep-wake[tw])) OR ((advanced[tw] OR delayed[tw]) AND sleep phase[tw]) OR ACKNOWLEDGMENTS (idiopathic[tw] AND chronic[tw] AND (sleep-onset[tw] OR sleep onset[tw]) AND insomnia[tw]) OR ((nonentrained[tw] OR The authors acknowledge Dr. Ilene Rosen for serving as a board liaison for this project and assuring that the guideline adhered to the standards set forth by the non-entrained[tw]) AND (sleep[tw] OR “sleep”[mesh] OR Board of Directors of American Academy of Sleep Medicine. The authors also ac- “Sleep Disorders”[Mesh:noexp] OR circadian[tw] OR knowledge the contributions of the following individuals: Christine Stepanski, MS, “Circadian Rhythm”[MeSH Terms])) OR ((DSPS[tw] OR for performing literature searches #1 and #2; and Michelle M. Tangredi, PhD, for ASPS[tw] OR early morning awakening[tw] OR (phase managing the project at the initial steps. The authors are grateful to the organiza- tions and individuals who provided their insights and suggestions during the public advance[All Fields] OR phase advance/delay[All Fields] OR comment period. Every comment was internally documented and addressed, and phase advanced[All Fields] OR phase advancement[All Fields]

1227 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

OR phase advancements[All Fields] OR phase advances[All ((owl[tw] OR owls[tw] OR lark[tw] OR larks[tw] OR Fields]) OR (phase delay[All Fields] OR phase delay/arrest[All morningness[tw] OR eveningness[tw] OR morning types[tw] Fields] OR phase delayed[All Fields] OR phase delaying[All OR morning type[tw] OR evening types[tw] OR evening Fields] OR phase delays[All Fields]) OR (phase shift[All Fields] type[tw]) AND (sleep[tw] OR “sleep”[mesh] OR “Sleep OR phase shifted[All Fields] OR phase shifter[All Fields] OR Disorders”[Mesh:noexp] OR circadian[tw] OR “Circadian phase shifter/modulator[All Fields] OR phase shifters[All Rhythm”[MeSH Terms]))) AND ((randomized controlled Fields] OR phase shifting[All Fields] OR phase shiftkeying[All trial[pt] OR controlled clinical trial[pt] OR randomized[tiab] Fields] OR phase shiftmeasurement[All Fields] OR phase OR placebo[tiab] OR “drug therapy”[Subheading] OR shifts[All Fields])) AND (sleep[tw] OR “sleep”[mesh] OR randomly[tiab] OR trial[tiab] OR groups[tiab]) NOT “Sleep Disorders”[Mesh:noexp] OR circadian[tw] OR (“animals”[MeSH Terms] NOT “humans”[MeSH Terms])) “Circadian Rhythm”[MeSH Terms])) OR ((free running[tw] AND (English[la] AND (“2006/10/01”[PDAT] : AND (disorder[tw] OR rhythm[tw])) AND (sleep[tw] OR “3000/12/31”[PDAT]) NOT (Comment[pt] OR Editorial[pt] OR “sleep”[mesh] OR “Sleep Disorders”[Mesh:noexp] OR News[pt] OR Newspaper Article[pt] OR Letter[pt] OR Case circadian[tw] OR “Circadian Rhythm”[MeSH Terms])) OR Reports[pt] OR Review[pt])))) NOT (((((circadian rhythm sleep ((dark therapy[tw] OR light therapy[tw] OR amber lenses[tw] disorders[tw] OR circadian rhythm sleep disorder[tw] OR OR Blue light[tw] OR bright light[tw] AND blue blockers[tw] “Sleep Disorders, Circadian Rhythm”[Mesh:noexp] OR OR blue-blocker[tw] OR blue blocker[tw] OR blue-blocker[tw] “chronobiology disorders”[tw] OR “Chronobiology OR eyewear[tw] OR Phototherapy[tw] OR Disorders”[Mesh:noexp] OR hypernychthemeral[tw] OR “Phototherapy”[Mesh]) AND (sleep[tw] OR “sleep”[mesh] OR circadian misalignment[tw] OR circadian dysregulation[tw])) “Sleep Disorders”[Mesh:noexp] OR circadian[tw] OR OR (((irregular[tw] OR non 24 hour[tw]) AND (sleep wake[tw] “Circadian Rhythm”[MeSH Terms])) OR “sleep phase OR sleep-wake[tw]))) OR (((advanced[tw] OR delayed[tw]) chronotherapy”[MeSH Terms] OR ((chronotherapy[tw] OR AND (sleep phase[tw]))) OR ((idiopathic[tw] AND chronic[tw] chronotype[tw] OR “Chronotherapy”[Mesh:noexp]) AND AND (sleep-onset[tw] OR sleep onset[tw]) AND insomnia[tw])) (sleep[tw] OR “sleep”[mesh] OR “Sleep OR (((nonentrained[tw] OR non-entrained[tw]) AND (sleep[tw] Disorders”[Mesh:noexp] OR circadian[tw] OR “Circadian OR “sleep”[mesh] OR “Sleep Disorders”[Mesh:noexp] OR Rhythm”[MeSH Terms])) OR ((melatonin receptor agonists[tw] circadian[tw] OR “Circadian Rhythm”[MeSH Terms]))) OR OR melatonin receptor agonist[tw] OR melatonin agonist[tw] (((DSPS[tw] OR ASPS[tw] OR phase advance* OR phase delay* OR melatonin agonists[tw] OR melatonin receptor OR phase shift*) AND (sleep[tw] OR “sleep”[mesh] OR “Sleep antagonists[tw] OR melatonin receptor antagonist[tw] OR Disorders”[Mesh:noexp] OR circadian[tw] OR “Circadian melatonin antagonist[tw] OR melatonin antagonists[tw] OR Rhythm”[MeSH Terms]))) OR (((free running[tw] AND “Receptors, Melatonin/drug effects”[Mesh] OR “Receptors, (disorder[tw] OR rhythm[tw])) AND (sleep[tw] OR Melatonin/agonists”[Mesh] OR “Receptors, Melatonin/ “sleep”[mesh] OR “Sleep Disorders”[Mesh:noexp] OR antagonists and inhibitors”[Mesh]) AND (sleep[tw] OR circadian[tw] OR “Circadian Rhythm”[MeSH Terms]))) OR “sleep”[mesh] OR “Sleep Disorders”[Mesh:noexp] OR (((dark therapy[tw] OR light therapy[tw] OR amber lenses[tw] circadian[tw] OR “Circadian Rhythm”[MeSH Terms])) OR OR Blue light[tw] OR bright light[tw] OR eyewear[tw] OR ((melatonin[tw] OR “melatonin”[MeSH Terms] OR Phototherapy[tw] OR “Phototherapy”[Mesh]) AND (sleep[tw] ramelteon[tw] OR “ramelteon”[Substance] OR tasimelteon[tw] OR “sleep”[mesh] OR “Sleep Disorders”[Mesh:noexp] OR OR “tasimelteon”[Supplementary Concept] OR circadian[tw] OR “Circadian Rhythm”[MeSH Terms]))) OR agomelatine[tw] OR “S 20098”[Supplementary Concept]) ((Sleep Phase Chronotherapy[Mesh])) OR (((chronotherapy[tw] AND (sleep[tw] OR “sleep”[mesh] OR “Sleep OR “Chronotherapy”[Mesh:noexp]) AND (sleep[tw] OR Disorders”[Mesh:noexp] OR circadian[tw] OR “Circadian “sleep”[mesh] OR “Sleep Disorders”[Mesh:noexp] OR Rhythm”[MeSH Terms])) OR ((dim light melatonin onset[tw] circadian[tw] OR “Circadian Rhythm”[MeSH Terms]))) OR OR DLMO[tw]) AND (sleep[tw] OR “sleep”[mesh] OR “Sleep (((melatonin receptor agonists[tw] OR melatonin receptor Disorders”[Mesh:noexp] OR circadian[tw] OR “Circadian agonist[tw] OR melatonin agonist[tw] OR melatonin Rhythm”[MeSH Terms])) OR ((Blindness[tw] OR blind agonists[tw] OR melatonin receptor antagonists[tw] OR person[tw] OR blind people[tw] OR blind subject[tw] OR blind melatonin receptor antagonist[tw] OR melatonin antagonist[tw] subjects[tw] OR blind patient[tw] OR blind patients[tw] OR OR melatonin antagonists[tw] OR melatonin receptor “Visually Impaired Persons”[Mesh] OR enucleated[tw] OR inhibitor[tw] OR melatonin receptor inhibitors[tw] OR “Blindness”[Mesh] OR nonsighted[tw] OR (visual melatonin inhibitor[tw] OR melatonin inhibitors[tw] OR impairement[All Fields] OR visual impairment[All Fields] OR “Receptors, Melatonin/drug effects”[Mesh] OR “Receptors, visual impairment/blindness[All Fields] OR visual Melatonin/agonists”[Mesh] OR “Receptors, Melatonin/ impairments[All Fields]) OR visually impaired[tw]) AND antagonists and inhibitors”[Mesh]) AND (sleep[tw] OR (sleep[tw] OR “sleep”[mesh] OR “Sleep “sleep”[mesh] OR “Sleep Disorders”[Mesh:noexp] OR Disorders”[Mesh:noexp] OR circadian[tw] OR “Circadian circadian[tw] OR “Circadian Rhythm”[MeSH Terms]))) OR Rhythm”[MeSH Terms])) OR ((“dementia”[MeSH Terms] OR (((melatonin[tw] OR “melatonin”[MeSH Terms] OR dementia[tw] OR “alzheimer disease”[MeSH Terms] OR ramelteon[tw] OR “ramelteon”[Substance] OR tasimelteon[tw] alzheimer[tw] OR alzheimers’s[tw]) AND (sleep[tw] OR OR “tasimelteon”[Supplementary Concept] OR “sleep”[mesh] OR “Sleep Disorders”[Mesh:noexp] OR agomelatine[tw] OR “S 20098”[Supplementary Concept]) circadian[tw] OR “Circadian Rhythm”[MeSH Terms])) OR AND (sleep[tw] OR “sleep”[mesh] OR “Sleep

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1228 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

Disorders”[Mesh:noexp] OR circadian[tw] OR “Circadian ‘chronotherapy’/exp) AND (‘sleep’:de,ab,ti OR ‘sleep’/exp Rhythm”[MeSH Terms]))) OR (((dim light melatonin onset[tw] OR ‘sleep disorder’/de OR ‘circadian’:de,ab,ti OR ‘circadian OR DLMO[tw]) AND (sleep[tw] OR “sleep”[mesh] OR “Sleep rhythm’/exp OR ‘sleep therapy’/exp)) OR ((‘melatonin receptor Disorders”[Mesh:noexp] OR circadian[tw] OR “Circadian agonists’:de,ab,ti OR ‘melatonin receptor agonist’:de,ab,ti Rhythm”[MeSH Terms]))) OR (((Blindness[tw] OR OR ‘melatonin agonist’:de,ab,ti OR ‘melatonin “Blindness”[Mesh] OR nonsighted[tw] OR (visual impair*) OR agonists’:de,ab,ti OR ‘melatonin receptor antagonists’:de,ab,ti (visually impaired[tw])) AND (sleep[tw] OR “sleep”[mesh] OR OR ‘melatonin receptor antagonist’:de,ab,ti OR ‘melatonin “Sleep Disorders”[Mesh:noexp] OR circadian[tw] OR antagonist’:de,ab,ti OR ‘melatonin antagonists’:de,ab,ti “Circadian Rhythm”[MeSH Terms]))) OR (((“dementia”[MeSH OR ‘melatonin receptor inhibitor’:de,ab,ti OR ‘melatonin Terms] OR dementia[tw] OR “alzheimer disease”[MeSH receptor inhibitors’:de,ab,ti OR ‘melatonin inhibitor’:de,ab,ti Terms] OR alzheimer[tw] OR alzheimers’s[tw]) AND (sleep[tw] OR ‘melatonin inhibitors’:de,ab,ti OR ‘melatonin OR “sleep”[mesh] OR “Sleep Disorders”[Mesh:noexp] OR receptor’/exp/dd_dt OR (‘melatonin receptor’/exp circadian[tw] OR “Circadian Rhythm”[MeSH Terms]))) OR AND ‘agonist’/exp)) AND (‘sleep’:de,ab,ti OR ‘sleep’/ (((owl[tw] OR owls[tw] OR lark[tw] OR larks[tw] OR exp OR ‘sleep disorder’/de OR ‘circadian’:de,ab,ti OR morningness[tw] OR eveningness[tw] OR morning types[tw] ‘circadian rhythm’/exp)) OR ((‘melatonin’:de,ab,ti OR OR morning type[tw] OR evening types[tw] OR evening ‘melatonin’/exp OR ‘ramelteon’:de,ab,ti OR ‘ramelteon’/ type[tw]) AND (sleep[tw] OR “sleep”[mesh] OR “Sleep exp OR ‘tasimelteon’:de,ab,ti OR ‘tasimelteon’/exp Disorders”[Mesh:noexp] OR circadian[tw] OR “Circadian OR ‘agomelatine’:de,ab,ti OR ‘agomelatine’/exp) AND Rhythm”[MeSH Terms]))))) AND ((randomized controlled (‘sleep’:de,ab,ti OR ‘sleep’/exp OR ‘sleep disorder’/de trial[pt] OR controlled clinical trial[pt] OR randomized[tiab] OR ‘circadian’:de,ab,ti OR ‘circadian rhythm’/exp)) OR OR placebo[tiab] OR “drug therapy”[Subheading] OR ((‘dim light melatonin onset’:de,ab,ti OR ‘DLMO’:de,ab,ti) randomly[tiab] OR trial[tiab] OR groups[tiab]) NOT AND (‘sleep’:de,ab,ti OR ‘sleep’/exp OR ‘sleep disorder’/ (“animals”[MeSH Terms] NOT “humans”[MeSH Terms])) de OR ‘circadian’:de,ab,ti OR ‘circadian rhythm’/ AND (English[la] AND (“2006/10/01”[PDAT] : exp)) OR ((‘Blindness’:de,ab,ti OR ‘blindness’/exp “3000/12/31”[PDAT]) NOT (Comment[pt] OR Editorial[pt] OR OR ‘blind person’:de,ab,ti OR ‘blind people’:de,ab,ti News[pt] OR Newspaper Article[pt] OR Letter[pt] OR Case OR ‘blind subject’:de,ab,ti OR ‘blind subjects’:de,ab,ti Reports[pt] OR Review[pt]))) OR ‘blind patient’:de,ab,ti OR ‘blind patients’:de,ab,ti OR ‘enucleated’:de,ab,ti OR ‘nonsighted’:de,ab,ti OR Terms for the Literature Search #2 in Embase Database (visual AND impair*) OR (‘visually impaired’:de,ab,ti)) (‘circadian rhythm sleep disorders’:de,ab,ti OR ‘circadian AND (‘sleep’:de,ab,ti OR ‘sleep’/exp OR ‘sleep disorder’/ rhythm sleep disorder’:de,ab,ti OR ‘circadian rhythm de OR ‘circadian’:de,ab,ti OR ‘circadian rhythm’/exp)) OR sleep disorder’/exp OR ‘chronobiology disorders’:de,ab,ti ((‘dementia’/exp OR ‘dementia’:de,ab,ti OR ‘Alzheimer OR ‘hypernychthemeral’:de,ab,ti OR ‘circadian disease’/exp OR ‘alzheimer’:de,ab,ti OR alzheimer*) misalignment’:de,ab,ti OR ‘circadian dysregulation’:de,ab,ti) AND (‘sleep’:de,ab,ti OR ‘sleep’/exp OR ‘sleep disorder’/ OR((‘advanced’:de,ab,ti OR ‘delayed’:de,ab,ti) AND de OR ‘circadian’:de,ab,ti OR ‘circadian rhythm’/exp)) (‘sleep phase’:de,ab,ti)) OR ((‘irregular’:de,ab,ti OR OR ((‘owl’:de,ab,ti OR ‘owls’:de,ab,ti OR ‘lark’:de,ab,ti ‘non 24 hour’:de,ab,ti) AND (‘sleep wake’:de,ab,ti OR ‘larks’:de,ab,ti OR ‘morningness’:de,ab,ti OR OR ‘sleep-wake’:de,ab,ti)) OR ((‘idiopathic’:de,ab,ti ‘eveningness’:de,ab,ti OR ‘morning types’:de,ab,ti OR AND ‘chronic’:de,ab,ti AND (‘sleep-onset’:de,ab,ti ‘morning type’:de,ab,ti OR ‘evening types’:de,ab,ti OR OR ‘sleep onset’:de,ab,ti) AND ‘insomnia’:de,ab,ti)) OR ‘evening type’:de,ab,ti) AND (‘sleep’:de,ab,ti OR ‘sleep’/exp ((‘nonentrained’:de,ab,ti OR ‘non-entrained’:de,ab,ti) OR ‘sleep disorder’/de OR ‘circadian’:de,ab,ti OR ‘circadian AND (‘sleep’:de,ab,ti OR ‘sleep’/exp OR ‘sleep disorder’/ rhythm’/exp)) AND ((random* OR factorial* OR crossover* de OR ‘circadian’:de,ab,ti OR ‘circadian rhythm’/exp)) OR OR cross NEXT/1 over* OR cross-over* OR placebo* OR ((‘DSPS’:de,ab,ti OR ‘ASPS’:de,ab,ti OR ‘early morning doubl* NEXT/1 blind* OR singl* NEXT/1 blind* OR assign* awakening’:de,ab,ti OR phase advance* OR phase delay* OR OR allocat* OR volunteer* OR ‘crossover procedure’/de OR phase shift*) AND (‘sleep’:de,ab,ti OR ‘sleep’/exp OR ‘sleep ‘double blind procedure’/de OR ‘randomized controlled trial’/ disorder’/de OR ‘circadian’:de,ab,ti OR ‘circadian rhythm’/ de OR ‘single blind procedure’/de) OR (‘clinical trial’/exp exp)) OR (((‘free running’:de,ab,ti AND (‘disorder’:de,ab,ti OR ‘counterbalance’:de,ab,ti OR ‘counterbalanced’:de,ab,ti OR ‘rhythm’:de,ab,ti)) AND (‘sleep’:de,ab,ti OR ‘sleep’/ OR ‘crossover’:de,ab,ti OR ‘trial’:de,ab,ti)) AND ((AND exp OR ‘sleep disorder’/de OR ‘circadian’:de,ab,ti OR [english]/lim AND [2006-2012]/py AND [embase]/lim NOT ‘circadian rhythm’/exp))) OR ((‘dark therapy’:de,ab,ti [medline]/lim NOT (‘editorial’:it OR ‘conference review’:it OR ‘light therapy’:de,ab,ti OR ‘amber lenses’:de,ab,ti OR OR ‘conference paper’:it OR ‘note’:it OR ‘conference ‘blue blocker’:de,ab,ti OR ‘blue-blocker’:de,ab,ti OR ‘blue abstract’:it OR ‘letter’:it OR ‘review’:it) NOT ([animals]/lim blockers’:de,ab,ti OR ‘blue-blockers’:de,ab,ti OR ‘Blue NOT [humans]/lim)) light’:de,ab,ti OR ‘bright light’:de,ab,ti OR ‘eyewear’:de,ab,ti OR ‘Phototherapy’:de,ab,ti OR ‘phototherapy’/exp) AND Terms for the Literature Search # 2 in PsychInfo (‘sleep’:de,ab,ti OR ‘sleep’/exp OR ‘sleep disorder’/de Database OR ‘circadian’:de,ab,ti OR ‘circadian rhythm’/exp)) OR ((TI “circadian rhythm sleep disorders” or TI “circadian ((‘chronotherapy’:de,ab,ti OR ‘chronotype’:de,ab,ti OR rhythm sleep disorder” or AB “circadian rhythm sleep

1229 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

disorders” or AB “circadian rhythm sleep disorder” or TI “melatonin receptor inhibitor” or AB “melatonin receptor TI “chronobiology disorders” or AB “chronobiology inhibitor” or TI “melatonin receptor inhibitors” or AB disorders” OR TI “chronobiology disorder” or AB “melatonin receptor inhibitors” or TI “melatonin inhibitor” “chronobiology disorder” or TI “hypernychthemeral” or AB or AB “melatonin inhibitor” or TI “melatonin inhibitors” or “hypernychthemeral” or TI “circadian misalignment” or AB AB “melatonin inhibitors”) AND (TI “sleep” or AB “sleep” “circadian misalignment” or TI “circadian dysregulation” or or DE “Sleep” or DE “Sleep Disorders” or TI “circadian” AB “circadian dysregulation”)) OR ((TI “advanced” or AB or AB “circadian” or DE “Human Biological Rhythms” or “advanced” or TI “delayed” or AB “delayed”) and (TI “sleep DE “Sleep Wake Cycle”)) OR ((TI “melatonin” OR AB phase” or AB “sleep phase”)) OR ((TI “irregular” or AB “melatonin” or DE “melatonin” OR TI “ramelteon” OR AB “irregular” or TI “non 24 hour” or AB “non 24 hour”) AND “ramelteon” OR TI “tasimelteon” OR AB “tasimelteon” OR (TI “sleep wake” or AB “sleep wake” or TI “sleep-wake” or TI “agomelatine” OR AB “agomelatine”) AND (TI “sleep” AB “sleep-wake”)) OR ((TI “idiopathic” or AB “idiopathic”) or AB “sleep” or DE “Sleep” or DE “Sleep Disorders” or and (TI “chronic” or AB “chronic” ) and (TI “sleep-onset” or TI “circadian” or AB “circadian” or DE “Human Biological AB “sleep-onset” or TI “sleep onset” or AB “sleep onset”) Rhythms” or DE “Sleep Wake Cycle”)) OR ((TI “dim light and (DE “Insomnia” or TI “insomnia” or AB “insomnia”)) melatonin onset” or AB “dim light melatonin onset” or TI OR ((TI “nonentrained” or AB “nonentrained” OR TI “non- “DLMO” or AB “DLMO”) and (TI “sleep” or AB “sleep” entrained” OR AB “non-entrained”) AND (TI “sleep” or or DE “Sleep” or DE “Sleep Disorders” or TI “circadian” AB “sleep” or DE “Sleep” or DE “Sleep Disorders” or TI or AB “circadian” or DE “Human Biological Rhythms” “circadian” or AB “circadian” or DE “Human Biological or DE “Sleep Wake Cycle”)) OR ((TI “Blindness” or AB Rhythms” or DE “Sleep Wake Cycle”)) OR ((TI “DSPS” “blindness” or TI “blind person” or AB “blind person” or or AB “DSPS” or TI “ASPS” or AB “ASPS” or TI “early TI “blind people” or AB “blind people” or TI “blind subject” morning awakening” or AB “early morning awakening” or or AB “blind subject” or TI “blind subjects” or AB “blind TI “phase advance*” or AB “phase advance*” or TI “phase subjects” or TI “blind patient” or AB “blind patient” or TI delay*” or AB “phase delay*” or TI “phase shift*” or AB “blind patients” or AB “blind patients” or TI “enucleated” “phase shift*”) and (TI “sleep” or AB “sleep” or DE “Sleep” or AB “enucleated” or TI “nonsighted” or AB “nonsighted” or DE “Sleep Disorders” or TI “circadian” or AB “circadian” or (TI “visual” and TI “impair*”) or (AB “visual” and or DE “Human Biological Rhythms” or DE “Sleep Wake TI “impair*”) or TI “visually impaired” or AB “visually Cycle”)) OR (((TI “free running” or AB “free running”) impaired” or DE “Vision Disorders” or DE “Balint’s and (TI “disorder” or AB “disorder” or TI “rhythm” or AB Syndrome” or DE “Blind” or DE “Eye Disorders” or DE “rhythm”)) and (TI “sleep” or AB “sleep” or DE “Sleep” or “Hemianopia”) and (TI “sleep” or AB “sleep” or DE “Sleep” DE “Sleep Disorders” or TI “circadian” or AB “circadian” or DE “Sleep Disorders” or TI “circadian” or AB “circadian” or DE “Human Biological Rhythms” or DE “Sleep Wake or DE “Human Biological Rhythms” or DE “Sleep Wake Cycle”))) OR ((TI “dark therapy” or AB “dark therapy” or Cycle”)) OR ((DE “Dementia” OR DE “AIDS Dementia TI “light therapy” or AB “light therapy” or TI “amber lenses” Complex” OR DE “Dementia with Lewy Bodies” OR DE or AB “amber lenses” or TI “blue blocker” or AB “blue “Presenile Dementia” OR DE “Semantic Dementia” OR blocker” or TI “blue-blocker” or AB “blue-blocker” or TI DE “Senile Dementia” OR DE “Vascular Dementia” or DE “blue blockers” or AB “blue blockers” or TI “blue-blockers” “Alzheimer’s Disease” OR TI “dementia” or AB “dementia” or AB “blue-blockers” or TI “Blue light” or AB “Blue light” or TI “Alzheimer*” or AB “Alzheimer*”) AND (TI “sleep” or TI “bright light” or AB “bright light” or TI “eyewear” or or AB “sleep” or DE “Sleep” or DE “Sleep Disorders” or AB “eyewear” or DE “Optical Aids” or TI “Phototherapy” TI “circadian” or AB “circadian” or DE “Human Biological or AB “Phototherapy” or DE “Phototherapy”) and (TI Rhythms” or DE “Sleep Wake Cycle”)) OR ((TI “owl” or AB “sleep” or AB “sleep” or DE “Sleep” or DE “Sleep Disorders” “owl” OR TI “owls” or AB “owls” or TI “lark” or AB “lark” or TI “circadian” or AB “circadian” or DE “Human OR TI “larks” or AB “larks” or TI “morningness” or AB Biological Rhythms” or DE “Sleep Wake Cycle”)) OR ((TI “morningness” or TI “eveningness” or AB “eveningness” or “chronotherapy” or AB “chronotherapy” or TI “chronotype” TI “morning types” or AB “morning types” or TI “morning or AB “chronotypes” or TI “chronotypes” or AB type” or AB “morning type” or TI “evening types” or AB “chronotypes” or TI “chronotherapy” or AB “chronotherapy”) “evening types” or TI “evening type” or AB “evening type” AND (TI “sleep” or AB “sleep” or DE “Sleep” or DE “Sleep or TM “Horne and Ostberg Morningness-Eveningness Disorders” or TI “circadian” or AB “circadian” or DE Questionnaire”) and (TI “sleep” or AB “sleep” or DE “Sleep” “Human Biological Rhythms” or DE “Sleep Wake Cycle”)) or DE “Sleep Disorders” or TI “circadian” or AB “circadian” OR ((TI “melatonin receptor agonists” or AB “melatonin or DE “Human Biological Rhythms” or DE “Sleep Wake receptor agonists” or TI “melatonin receptor agonist” or AB Cycle”)) AND ((randomized controlled trial[pt] OR controlled “melatonin receptor agonist” or TI “melatonin agonist” or clinical trial[pt] OR randomized[tiab] OR placebo[tiab] AB “melatonin agonist” or TI “melatonin agonists” or AB OR “drug therapy”[Subheading] OR randomly[tiab] OR “melatonin agonists” OR TI “melatonin receptor antagonists” trial[tiab] OR groups[tiab]) NOT (“animals”[MeSH Terms] or AB “melatonin receptor antagonists” or TI “melatonin NOT “humans”[MeSH Terms])) AND ((English[la] AND receptor antagonist” or AB “melatonin receptor antagonist” (“2006/10/01”[PDAT] : “3000/12/31”[PDAT]) NOT (Comment or TI “melatonin antagonist” or AB “melatonin antagonist” or [pt] OR Editorial [pt] OR News [pt] OR Newspaper Article TI “melatonin antagonists” or AB “melatonin antagonists” or [pt] OR Letter [pt] OR Case Reports [pt] OR Review [pt]))

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1230 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders Terms for the Literature Search for Harms/Adverse Terms for the Literature Search for Harms/Adverse Effects of light Treatment (Pubmed Database) Effects of Melatonin Treatment (Pubmed Database) (harm* OR “side effect*” OR “adverse effect*”) and (Light* (harm* OR “side effect*” OR “adverse effect*”) and OR “light therapy*” OR “light exposure*” OR “Light (melatonin*) treatment*”) and “sleep” Limits Used for the Searches of Harms/Adverse Terms for the Literature Search for Harms/Adverse Effects of Treatments Effects of Treatment with Hypnotics (Pubmed Article types: meta-analysis, systematic review Database) Species: humans (harm* OR “side effect*” OR “adverse effect*”) and Languages: English (hypnotic*)

Summary of Findings Tables (Tables S1–S12)

Table S1—Light treatment compared to placebo for ASWPD in adults.

Bibliography: APalmer et al., 2003; BCampbell et al., 1993. 1CI of the absolute effect crosses the clinical significance threshold (defined by the TF) on both sides of the “no effect” line. 2CI of the absolute effect crosses the clinical significance threshold defined by the TF.3 CI of the absolute effect crosses the clinical significance threshold (defined by the TF) and the “no effect” line. aMT6, 6-sulfatoxymelatonin (urinary metabolite of melatonin); TST, total sleep time; ISL,

initial sleep latency; SOT, sleep onset time; SOffT, sleep offset time; CBTMin, core body temperature minimum; CI, confidence interval; min, minutes; PSG, polysomnography; TF, task force.

1231 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

Table S2—Melatonin treatment compared to placebo for DSWPD in adults with and without depression.

Bibliography: AKayumov et al., 2001; BMundey et al., 2005; CRahman et al., 2010. *In studies with crossover design, where more than one measurements were taken on the same patients, all calculations were made based on the number of measurements. 1High level of heterogeneity. 2CI of the absolute effect crosses the clinical significance threshold defined by the TF.3 CI of the absolute effect crosses the clinical significance threshold (defined by the TF) and the “no effect” line. 4CI of the absolute effect crosses the clinical significance threshold (defined by the TF) on both sides of the “no effect” line. TST, total sleep time; ISL, initial sleep latency; DLMO, dim light melatonin onset; SOT, sleep onset time; SOffT, sleep offset time; CI, confidence interval; PSG, polysomnography; TF, task force; min, minutes.

Table S3—Melatonin treatment compared to placebo for DSWPD in children/adolescents with no comorbidities.

Bibliography: van Geijlswijk et al., 2010. 1CI of the absolute effect crosses the clinical significance threshold defined by the TF on both sides of the “no effect” line. 2CI of the absolute effect crosses the clinical significance threshold (defined by the TF) and the “no effect” line. 3CI of the absolute effect crosses the clinical significance threshold defined by the TF. DLMO, dim light melatonin onset; ISL, initial sleep latency; SOT, sleep onset time; CI, confidence interval; mg, milligrams; min, minutes; kg, kilograms; TF, task force.

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1232 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

Table S4—Melatonin treatment compared to placebo for DSWPD in children/adolescents comorbid with psychiatric conditions.

Bibliography: ASmits et al., 2001; Bvan der Heijden et al., 2007. 1CI of the absolute effect crosses the “no effect” line and the clinical significance threshold defined by the TF. 2CI of the absolute effect crosses the clinical significance threshold defined by the TF. 3CI of the absolute effect crosses the clinical significance threshold (defined by the TF) on both sides of the “no effect” line. TST, total sleep time; ISL, initial sleep latency; DLMO, dim light melatonin onset; SOT, sleep onset time; SOffT, sleep offset time; CI, confidence interval; TF, task force; min, minutes.

Table S5—Vitamin B12 compared to placebo for DSWPD in adults.

Bibliography: Okawa et al., 1997. 1CI of the absolute effect crosses the clinical significance threshold defined by the TF. TST, total sleep time; SOT, sleep onset time; CI, confidence interval; TF, task force; min, minutes.

1233 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

Table S6—Light/combination treatment compared to controls for DSWPD in adults.

Bibliography: ACole et al., 2002; BLack, Bramwell et al., 2007. 1This is a short paper with lack of detail in methods and results; a lot of information is missing. 2Confidence interval of the absolute effect crosses the clinical significance threshold (defined by the TF) on both sides of the “no effect line.” 3Confidence interval of the absolute effect crosses the “no effect” line. 4Confidence interval of the absolute effect crosses the clinical significance threshold defined by the TF. TST, total sleep time; SOT, sleep onset time; SOffT, sleep offset time; aMT6, 6-sulfatoxymelatonin (urinary metabolite of melatonin); CI, confidence interval; TF, task force; min, minutes.

Table S7—Light/combination treatment (multicomponent behavioral interventions + light) compared to controls for DSWPD in children/adolescents.

Bibliography: Gradisar et al., 2011. 1Study was not blinded, and an intention to treat analysis was NOT utilized. 2CI of the absolute effect does not reach clinical significance threshold defined by the TF. TST, total sleep time; ISL, initial sleep latency; SOT, sleep onset time; SOffT, sleep offset time; CI, confidence interval; TF, task force; min, minutes.

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1234 Treatment of Intrinsic Circadian Rhythm Sleep-Wake Disorders

Table S8—Melatonin treatment compared to placebo for N24SWD in blind adults.

Bibliography: Lockley et al., 2000; Sack et al., 2000; Hack et al., 2003. CI, confidence interval; OR, odds ratio.

Table S9—Light treatment compared to placebo for ISWRD in elderly with dementia.

Bibliography: ADowling et al., 2008; BMishima et al., 1994. 1A. No intention-to-treat principle: only completed cases were analyzed. B. Selective reporting: dropouts are not described and some actigraphy variables are not reported (sleep onset, offset, sleep efficiency etc.).2 CI of the absolute value crosses the clinical significance threshold (defined by the TF) on both sides of the “no effect” line. 3A. No blinding, B. No intention-to-treat principle observed. C. Role of control group is vaguely described. 3CI of the absolute effect crosses the clinical significance threshold (defined by the TF) on both side of the “no effect” line. TST, total sleep time; CI, confidence interval; TF, task force; min, minutes.

Table S10—Melatonin treatment compared to placebo for ISWRD in elderly with dementia.

Bibliography: Serfaty et al., 2002. 1CI of the absolute effect crosses the clinical significance threshold (defined by the TF) on both sides of “no effect” line. TST, total sleep time; CI, confidence interval; TF, task force; min, minute.

Table S11—Melatonin treatment compared to placebo for ISWRD in children/adolescents with neurologic disorders.

Bibliography: Wright et al., 2011. 1CI of the absolute effect crosses the clinical significance threshold defined by the TF. TST, total sleep time; ISL, initial sleep latency; CI, confidence interval; TF, task force; min, minutes.

1235 Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 RR Auger, HJ Burgess, JS Emens et al.

Table S12—Comparison of combination of light therapy and melatonin/placebo for ISWRD in elderly with dementia.

Bibliography: Dowling et al., 2008. 1A. No intention-to-treat principle: only completed cases were analyzed. B. Selective reporting: dropouts are not described and some actigraphy variables are not reported (sleep onset,/offset, etc.). 2CI of the absolute effect crosses the clinical significance threshold (defined by the TF) on both sides of the “no effect” line. PSWS, prescribed sleep-wake scheduling; TST, total sleep time; CI, confidence interval; TF, task force; min, minutes.

Journal of Clinical Sleep Medicine, Vol. 11, No. 10, 2015 1236